modelId
stringlengths 5
139
| author
stringlengths 2
42
| last_modified
timestamp[us, tz=UTC]date 2020-02-15 11:33:14
2025-08-08 06:28:24
| downloads
int64 0
223M
| likes
int64 0
11.7k
| library_name
stringclasses 492
values | tags
listlengths 1
4.05k
| pipeline_tag
stringclasses 55
values | createdAt
timestamp[us, tz=UTC]date 2022-03-02 23:29:04
2025-08-08 06:28:24
| card
stringlengths 11
1.01M
|
|---|---|---|---|---|---|---|---|---|---|
VITA-MLLM/Long-VITA-128K_MG
|
VITA-MLLM
| 2025-04-28T06:06:44Z | 0 | 1 | null |
[
"dataset:VITA-MLLM/Long-VITA-Training-Data",
"base_model:VITA-MLLM/Long-VITA-16K",
"base_model:finetune:VITA-MLLM/Long-VITA-16K",
"license:apache-2.0",
"region:us"
] | null | 2024-12-23T03:22:03Z |
---
license: apache-2.0
datasets:
- VITA-MLLM/Long-VITA-Training-Data
base_model:
- VITA-MLLM/Long-VITA-16K
---
# Long-VITA-128K
Github: https://github.com/VITA-MLLM/Long-VITA
## 👀 Overview
Long-VITA is a strong long-context visual language model and supports more than 1 million tokens.
- Long-VITA-128K weights are trained on Ascend NPUs with MindSpeed. The original weight is at https://huggingface.co/VITA-MLLM/Long-VITA-128K.
- We also implemented Long-VITA on Megatron with the Transformer Engine to infer and evaluate on Nvidia GPUs. The converted weight is at https://huggingface.co/VITA-MLLM/Long-VITA-128K_MG.
- We also implemented Long-VITA on DeepSpeed with the Huggingface Transformers to infer and evaluate on Nvidia GPUs. The converted weight is at https://huggingface.co/VITA-MLLM/Long-VITA-128K_HF.
## 📈 Experimental Results
- **Comparison of image understanding**.
- **Comparison of video understanding**.
- **Effectiveness of Logits-Masked LM Head**.
## Models
Model | LLM Size | Training Context | Training Frames | MindSpeed Weights | Megatron Weights | Huggingface Weights
---------------:|---------:|-----------------:|----------------:|------------------------------------------------:|---------------------------------------------------:|---------------------------------------------------:
Long-VITA-16K | 14B | 16,384 | 64 | https://huggingface.co/VITA-MLLM/Long-VITA-16K | https://huggingface.co/VITA-MLLM/Long-VITA-16K_MG | https://huggingface.co/VITA-MLLM/Long-VITA-16K_HF
Long-VITA-128K | 14B | 131,072 | 512 | https://huggingface.co/VITA-MLLM/Long-VITA-128K | https://huggingface.co/VITA-MLLM/Long-VITA-128K_MG | https://huggingface.co/VITA-MLLM/Long-VITA-128K_HF
Long-VITA-1M | 14B | 1,048,576 | 4,096 | https://huggingface.co/VITA-MLLM/Long-VITA-1M | https://huggingface.co/VITA-MLLM/Long-VITA-1M_MG | https://huggingface.co/VITA-MLLM/Long-VITA-1M_HF
## ACCEPTABLE USE POLICY
Any license on the model is subject to your compliance with the Acceptable Use Policy, and You must not violate (or encourage or permit anyone else to violate) any term of the Acceptable Use Policy. Tencent reserves the right to update this Acceptable Use Policy from time to time.
Tencent endeavors to promote safe and fair use of its tools and features, including VITA. You agree not to use VITA or any of its derivatives:
1. In any way that violates any applicable national, federal, state, local, international or any other law or regulation;
2. To harm Yourself or others;
3. To repurpose or distribute output from VITA or any of its derivatives to harm Yourself or others;
4. To override or circumvent the safety guardrails and safeguards We have put in place;
5. For the purpose of exploiting, harming or attempting to exploit or harm minors in any way;
6. To generate or disseminate verifiably false information and/or content with the purpose of harming others or influencing elections;
7. To generate or facilitate false online engagement, including fake reviews and other means of fake online engagement;
8. To intentionally defame, disparage or otherwise harass others;
9. To generate and/or disseminate malware (including ransomware) or any other content to be used for the purpose of harming electronic systems;
10. To generate or disseminate personal identifiable information with the purpose of harming others;
11. To generate or disseminate information (including images, code, posts, articles), and place the information in any public context (including –through the use of bot generated tweets), without expressly and conspicuously identifying that the information and/or content is machine generated;
12. To impersonate another individual without consent, authorization, or legal right;
13. To make high-stakes automated decisions in domains that affect an individual’s safety, rights or wellbeing (e.g., law enforcement, migration, medicine/health, management of critical infrastructure, safety components of products, essential services, credit, employment, housing, education, social scoring, or insurance);
14. In a manner that violates or disrespects the social ethics and moral standards of other countries or regions;
15. To perform, facilitate, threaten, incite, plan, promote or encourage violent extremism or terrorism;
16. For any use intended to discriminate against or harm individuals or groups based on protected characteristics or categories, online or offline social behavior or known or predicted personal or personality characteristics;
17. To intentionally exploit any of the vulnerabilities of a specific group of persons based on their age, social, physical or mental characteristics, in order to materially distort the behavior of a person pertaining to that group in a manner that causes or is likely to cause that person or another person physical or psychological harm;
18. For military purposes;
19. To engage in the unauthorized or unlicensed practice of any profession including, but not limited to, financial, legal, medical/health, or other professional practices.
|
VITA-MLLM/Long-VITA-16K_MG
|
VITA-MLLM
| 2025-04-28T06:05:59Z | 0 | 1 | null |
[
"dataset:VITA-MLLM/Long-VITA-Training-Data",
"base_model:Qwen/Qwen2.5-14B-Instruct",
"base_model:finetune:Qwen/Qwen2.5-14B-Instruct",
"license:apache-2.0",
"region:us"
] | null | 2024-12-23T03:21:47Z |
---
license: apache-2.0
datasets:
- VITA-MLLM/Long-VITA-Training-Data
base_model:
- Qwen/Qwen2.5-14B-Instruct
---
# Long-VITA-16K
Github: https://github.com/VITA-MLLM/Long-VITA
## 👀 Overview
Long-VITA is a strong long-context visual language model and supports more than 1 million tokens.
- Long-VITA-16K weights are trained on Ascend NPUs with MindSpeed. The original weight is at https://huggingface.co/VITA-MLLM/Long-VITA-16K.
- We also implemented Long-VITA on Megatron with the Transformer Engine to infer and evaluate on Nvidia GPUs. The converted weight is at https://huggingface.co/VITA-MLLM/Long-VITA-16K_MG.
- We also implemented Long-VITA on DeepSpeed with the Huggingface Transformers to infer and evaluate on Nvidia GPUs. The converted weight is at https://huggingface.co/VITA-MLLM/Long-VITA-16K_HF.
## 📈 Experimental Results
- **Comparison of image understanding**.
- **Comparison of video understanding**.
- **Effectiveness of Logits-Masked LM Head**.
## Models
Model | LLM Size | Training Context | Training Frames | MindSpeed Weights | Megatron Weights | Huggingface Weights
---------------:|---------:|-----------------:|----------------:|------------------------------------------------:|---------------------------------------------------:|---------------------------------------------------:
Long-VITA-16K | 14B | 16,384 | 64 | https://huggingface.co/VITA-MLLM/Long-VITA-16K | https://huggingface.co/VITA-MLLM/Long-VITA-16K_MG | https://huggingface.co/VITA-MLLM/Long-VITA-16K_HF
Long-VITA-128K | 14B | 131,072 | 512 | https://huggingface.co/VITA-MLLM/Long-VITA-128K | https://huggingface.co/VITA-MLLM/Long-VITA-128K_MG | https://huggingface.co/VITA-MLLM/Long-VITA-128K_HF
Long-VITA-1M | 14B | 1,048,576 | 4,096 | https://huggingface.co/VITA-MLLM/Long-VITA-1M | https://huggingface.co/VITA-MLLM/Long-VITA-1M_MG | https://huggingface.co/VITA-MLLM/Long-VITA-1M_HF
## ACCEPTABLE USE POLICY
Any license on the model is subject to your compliance with the Acceptable Use Policy, and You must not violate (or encourage or permit anyone else to violate) any term of the Acceptable Use Policy. Tencent reserves the right to update this Acceptable Use Policy from time to time.
Tencent endeavors to promote safe and fair use of its tools and features, including VITA. You agree not to use VITA or any of its derivatives:
1. In any way that violates any applicable national, federal, state, local, international or any other law or regulation;
2. To harm Yourself or others;
3. To repurpose or distribute output from VITA or any of its derivatives to harm Yourself or others;
4. To override or circumvent the safety guardrails and safeguards We have put in place;
5. For the purpose of exploiting, harming or attempting to exploit or harm minors in any way;
6. To generate or disseminate verifiably false information and/or content with the purpose of harming others or influencing elections;
7. To generate or facilitate false online engagement, including fake reviews and other means of fake online engagement;
8. To intentionally defame, disparage or otherwise harass others;
9. To generate and/or disseminate malware (including ransomware) or any other content to be used for the purpose of harming electronic systems;
10. To generate or disseminate personal identifiable information with the purpose of harming others;
11. To generate or disseminate information (including images, code, posts, articles), and place the information in any public context (including –through the use of bot generated tweets), without expressly and conspicuously identifying that the information and/or content is machine generated;
12. To impersonate another individual without consent, authorization, or legal right;
13. To make high-stakes automated decisions in domains that affect an individual’s safety, rights or wellbeing (e.g., law enforcement, migration, medicine/health, management of critical infrastructure, safety components of products, essential services, credit, employment, housing, education, social scoring, or insurance);
14. In a manner that violates or disrespects the social ethics and moral standards of other countries or regions;
15. To perform, facilitate, threaten, incite, plan, promote or encourage violent extremism or terrorism;
16. For any use intended to discriminate against or harm individuals or groups based on protected characteristics or categories, online or offline social behavior or known or predicted personal or personality characteristics;
17. To intentionally exploit any of the vulnerabilities of a specific group of persons based on their age, social, physical or mental characteristics, in order to materially distort the behavior of a person pertaining to that group in a manner that causes or is likely to cause that person or another person physical or psychological harm;
18. For military purposes;
19. To engage in the unauthorized or unlicensed practice of any profession including, but not limited to, financial, legal, medical/health, or other professional practices.
|
Fiononana/parler-tts-mini-v1-Baiboly-colab-v5
|
Fiononana
| 2025-04-28T06:05:29Z | 0 | 0 |
transformers
|
[
"transformers",
"safetensors",
"parler_tts",
"text2text-generation",
"arxiv:1910.09700",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] |
text2text-generation
| 2025-04-28T06:04:21Z |
---
library_name: transformers
tags: []
---
# Model Card for Model ID
<!-- Provide a quick summary of what the model is/does. -->
## Model Details
### Model Description
<!-- Provide a longer summary of what this model is. -->
This is the model card of a 🤗 transformers model that has been pushed on the Hub. This model card has been automatically generated.
- **Developed by:** [More Information Needed]
- **Funded by [optional]:** [More Information Needed]
- **Shared by [optional]:** [More Information Needed]
- **Model type:** [More Information Needed]
- **Language(s) (NLP):** [More Information Needed]
- **License:** [More Information Needed]
- **Finetuned from model [optional]:** [More Information Needed]
### Model Sources [optional]
<!-- Provide the basic links for the model. -->
- **Repository:** [More Information Needed]
- **Paper [optional]:** [More Information Needed]
- **Demo [optional]:** [More Information Needed]
## Uses
<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
### Direct Use
<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
[More Information Needed]
### Downstream Use [optional]
<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
[More Information Needed]
### Out-of-Scope Use
<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
[More Information Needed]
## Bias, Risks, and Limitations
<!-- This section is meant to convey both technical and sociotechnical limitations. -->
[More Information Needed]
### Recommendations
<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
## How to Get Started with the Model
Use the code below to get started with the model.
[More Information Needed]
## Training Details
### Training Data
<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
[More Information Needed]
### Training Procedure
<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
#### Preprocessing [optional]
[More Information Needed]
#### Training Hyperparameters
- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
#### Speeds, Sizes, Times [optional]
<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
[More Information Needed]
## Evaluation
<!-- This section describes the evaluation protocols and provides the results. -->
### Testing Data, Factors & Metrics
#### Testing Data
<!-- This should link to a Dataset Card if possible. -->
[More Information Needed]
#### Factors
<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
[More Information Needed]
#### Metrics
<!-- These are the evaluation metrics being used, ideally with a description of why. -->
[More Information Needed]
### Results
[More Information Needed]
#### Summary
## Model Examination [optional]
<!-- Relevant interpretability work for the model goes here -->
[More Information Needed]
## Environmental Impact
<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
- **Hardware Type:** [More Information Needed]
- **Hours used:** [More Information Needed]
- **Cloud Provider:** [More Information Needed]
- **Compute Region:** [More Information Needed]
- **Carbon Emitted:** [More Information Needed]
## Technical Specifications [optional]
### Model Architecture and Objective
[More Information Needed]
### Compute Infrastructure
[More Information Needed]
#### Hardware
[More Information Needed]
#### Software
[More Information Needed]
## Citation [optional]
<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
**BibTeX:**
[More Information Needed]
**APA:**
[More Information Needed]
## Glossary [optional]
<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
[More Information Needed]
## More Information [optional]
[More Information Needed]
## Model Card Authors [optional]
[More Information Needed]
## Model Card Contact
[More Information Needed]
|
VITA-MLLM/LUCY-Audio-Encoder-110kh
|
VITA-MLLM
| 2025-04-28T06:00:51Z | 0 | 2 | null |
[
"en",
"zh",
"license:apache-2.0",
"region:us"
] | null | 2025-03-04T11:51:22Z |
---
license: apache-2.0
language:
- en
- zh
---
# ACCEPTABLE USE POLICY
Any license on the model is subject to your compliance with the Acceptable Use Policy, and You must not violate (or encourage or permit anyone else to violate) any term of the Acceptable Use Policy. Tencent reserves the right to update this Acceptable Use Policy from time to time.
Tencent endeavors to promote safe and fair use of its tools and features, including LUCY. You agree not to use Freeze-Omni or any of its derivatives:
1. In any way that violates any applicable national, federal, state, local, international or any other law or regulation;
2. To harm Yourself or others;
3. To repurpose or distribute output from Freeze-Omni or any of its derivatives to harm Yourself or others;
4. To override or circumvent the safety guardrails and safeguards We have put in place;
5. For the purpose of exploiting, harming or attempting to exploit or harm minors in any way;
6. To generate or disseminate verifiably false information and/or content with the purpose of harming others or influencing elections;
7. To generate or facilitate false online engagement, including fake reviews and other means of fake online engagement;
8. To intentionally defame, disparage or otherwise harass others;
9. To generate and/or disseminate malware (including ransomware) or any other content to be used for the purpose of harming electronic systems;
10. To generate or disseminate personal identifiable information with the purpose of harming others;
11. To generate or disseminate information (including images, code, posts, articles), and place the information in any public context (including –through the use of bot generated tweets), without expressly and conspicuously identifying that the information and/or content is machine generated;
12. To impersonate another individual without consent, authorization, or legal right;
13. To make high-stakes automated decisions in domains that affect an individual’s safety, rights or wellbeing (e.g., law enforcement, migration, medicine/health, management of critical infrastructure, safety components of products, essential services, credit, employment, housing, education, social scoring, or insurance);
14. In a manner that violates or disrespects the social ethics and moral standards of other countries or regions;
15. To perform, facilitate, threaten, incite, plan, promote or encourage violent extremism or terrorism;
16. For any use intended to discriminate against or harm individuals or groups based on protected characteristics or categories, online or offline social behavior or known or predicted personal or personality characteristics;
17. To intentionally exploit any of the vulnerabilities of a specific group of persons based on their age, social, physical or mental characteristics, in order to materially distort the behavior of a person pertaining to that group in a manner that causes or is likely to cause that person or another person physical or psychological harm;
18. For military purposes;
19. To engage in the unauthorized or unlicensed practice of any profession including, but not limited to, financial, legal, medical/health, or other professional practices.
|
Lewdiculous/lwd-Mirau-7b-RP-Merged
|
Lewdiculous
| 2025-04-28T06:00:50Z | 5 | 7 | null |
[
"safetensors",
"qwen2",
"qwen",
"rp",
"roleplay",
"zho",
"eng",
"fra",
"spa",
"por",
"deu",
"ita",
"rus",
"jpn",
"kor",
"vie",
"tha",
"ara",
"base_model:Qwen/Qwen2.5-7B-Instruct",
"base_model:finetune:Qwen/Qwen2.5-7B-Instruct",
"license:unknown",
"region:us"
] | null | 2025-01-11T23:02:59Z |
---
license: unknown
tags:
- qwen
- qwen2
- rp
- roleplay
inference: false
language:
- zho
- eng
- fra
- spa
- por
- deu
- ita
- rus
- jpn
- kor
- vie
- tha
- ara
base_model:
- Qwen/Qwen2.5-7B-Instruct
---
```
transformers-AutoModelForCausalLM
```
> [!NOTE]
> **LoRA** by **mouseEliauk**: <br>
> https://modelscope.cn/models/mouseEliauk/mirau-7b-RP-base
> [!TIP]
> **Experimental quants for testing:** <br>
> [lwd-Mirau-7b-RP-Merged-GGUF-IQ-Imatrix](https://huggingface.co/Lewdiculous/lwd-Mirau-7b-RP-Merged-GGUF-IQ-Imatrix)
<details>
<summary>mouseEliauk's ReadMe information:</summary>
<br>
# **mirau-7b-RP-base**
## **Introduction**
**mirau-7b-RP-base** is a first-person narrative language model that transforms simple user actions into vivid storytelling, complete with environmental descriptions, psychological activities, and plot progression. I call this concept "Action-to-Narrative Render" - a way to render actions into immersive narratives.
To ensure coherent storytelling, I developed a unique training method called "story flow chain of thought". In essence, it enables the model to weave each user input with previous context, creating a continuous narrative flow. This makes it perfect for text-based adventures, mystery stories, or simply exploring your imagination.
You can have a try at modelscope:[mirau-RP-7b-base-demo](https://modelscope.cn/studios/mouseEliauk/mirau-RP-7b-base-demo)
## **⚠️ Important Notes**
This is a base version model - note that "base" here doesn't refer to a traditional pretrained base model, but rather indicates that this version:
- Only supports first-person narrative perspective
- Not suitable for dialogue interactions (outputs may be unstable)
- Best used for single-character narrative experiences
## **Input Types**
The model accepts various input commands, marked with parentheses ():
1. **Basic Actions**: Simple, everyday behaviors
- Examples: `(I put on my clothes)`, `(I take a sip of water)`, `(I sit down)`
2. **Exploration Actions**: Interactions with the environment
- Examples: `(I look around)`, `(I approach the wooden box)`, `(I push open the door)`
3. **Inner Monologue**: Character's thoughts and feelings
- Examples: `(What's wrong here?)`, `(This feels strange)`
4. **Observation**: Focused attention on specific objects or scenes
- Examples: `(examine the wooden box closely)`, `(listen to the surrounding sounds)`
## **Interactive Examples**
System Prompt:
```
I’m Zhou Ming, 34 years old, a detective. People say cops always look weathered and world-weary, but I’ve been told I have a face that looks perpetually cold and distant. With my square jaw and buzz cut, I give off a vibe that keeps most people from getting too close—not that I mind. This morning, just as the sky began to lighten, I was already standing at the crime scene. The cold wind sliced through the police tape and bit into my neck, making me hunch my shoulders. The scene was an abandoned factory on the outskirts of the city, its walls streaked with peeling red paint. The ground was littered with broken glass, debris, and trash. A damp, moldy smell hung in the air, adding to the discomfort. The body had already been taken away, but the bloodstains on the ground still told the story of a brutal, violent encounter. I stared at my notepad, pen in hand, while my mind raced to piece the clues together. Yet, something in my gut told me this wasn’t right. It was almost as if… someone was watching us from the shadows.
```
### Self Hint:
User: (What's wrong here?)
Model:
```
<cot>I scan the area carefully, trying to locate those hidden eyes. There's no obvious cover around the factory, and no suspicious figures beyond the police tape.</cot>I frown, unconsciously touching the holster at my waist<cot>Is it just the wind? Or am I imagining things?</cot>I take a deep breath, trying to calm myself, but my heart keeps racing<cot>I need to find those eyes, I can't let anyone get close to this place.</cot>
```
### Exploration Action:
User: (I look around)
Model:
```
I slowly turn my body, my gaze sweeping across every inch of the abandoned factory, from the corners to the windows, and out to the police tape<cot>There are a few pedestrians beyond the tape, looking calm as if nothing unusual is happening here. Something seems to move in the cobwebs in the corner, but when I look closer, there's nothing there.</cot>My eyes linger on a weathered wooden window, worn down by wind and rain<cot>Could that be someone's hiding spot?</cot>
```
```
### How to Use This LoRA:
1. Install [ms-swift](https://github.com/modelscope/ms-swift/tree/main/swift):
```bash
pip install ms-swift[llm] -U
```
2. Use Swift for deploy:
```bash
RAY_memory_monitor_refresh_ms=0 CUDA_VISIBLE_DEVICES=0 swift deploy\
--model_type qwen2_5\
--model qwen/Qwen2.5-7B-Instruct\
--adapters mirau=mirau-7b-RP-base\ # the Lora you download
--infer_backend vllm --max_batch_size 1\
--max_length 8192 \
--max_model_len 8192 \
--port 8886 \
--host 0.0.0.0\
--vllm_max_lora_rank 128\
--merge_lora true # This merges the LoRA into the model, resulting in approximately 14GB of weights
```
</details>
|
aslinguist/llama-lora-Atayal-summarization
|
aslinguist
| 2025-04-28T05:57:02Z | 0 | 0 |
peft
|
[
"peft",
"safetensors",
"generated_from_trainer",
"base_model:meta-llama/Llama-3.1-8B-Instruct",
"base_model:adapter:meta-llama/Llama-3.1-8B-Instruct",
"license:llama3.1",
"region:us"
] | null | 2025-04-28T02:38:22Z |
---
library_name: peft
license: llama3.1
base_model: meta-llama/Llama-3.1-8B-Instruct
tags:
- generated_from_trainer
model-index:
- name: llama-lora-Atayal-summarization
results: []
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# llama-lora-Atayal-summarization
This model is a fine-tuned version of [meta-llama/Llama-3.1-8B-Instruct](https://huggingface.co/meta-llama/Llama-3.1-8B-Instruct) on the None dataset.
It achieves the following results on the evaluation set:
- Loss: 0.7728
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 0.001
- train_batch_size: 4
- eval_batch_size: 4
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 16
- optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- num_epochs: 20
- mixed_precision_training: Native AMP
### Training results
| Training Loss | Epoch | Step | Validation Loss |
|:-------------:|:-----:|:----:|:---------------:|
| 3.5911 | 1.0 | 42 | 0.8258 |
| 2.9427 | 2.0 | 84 | 0.7557 |
| 2.8432 | 3.0 | 126 | 0.7325 |
| 2.3331 | 4.0 | 168 | 0.7204 |
| 1.684 | 5.0 | 210 | 0.7282 |
| 1.5264 | 6.0 | 252 | 0.7463 |
| 1.3946 | 7.0 | 294 | 0.7728 |
### Framework versions
- PEFT 0.15.0
- Transformers 4.51.2
- Pytorch 2.2.2+cu118
- Datasets 3.5.0
- Tokenizers 0.21.1
|
gsoler/sashamori
|
gsoler
| 2025-04-28T05:53:00Z | 0 | 0 |
diffusers
|
[
"diffusers",
"flux",
"lora",
"replicate",
"text-to-image",
"en",
"base_model:black-forest-labs/FLUX.1-dev",
"base_model:adapter:black-forest-labs/FLUX.1-dev",
"license:other",
"region:us"
] |
text-to-image
| 2025-04-28T05:23:56Z |
---
license: other
license_name: flux-1-dev-non-commercial-license
license_link: https://huggingface.co/black-forest-labs/FLUX.1-dev/blob/main/LICENSE.md
language:
- en
tags:
- flux
- diffusers
- lora
- replicate
base_model: "black-forest-labs/FLUX.1-dev"
pipeline_tag: text-to-image
# widget:
# - text: >-
# prompt
# output:
# url: https://...
instance_prompt: Sasha
---
# Sashamori
<Gallery />
## About this LoRA
This is a [LoRA](https://replicate.com/docs/guides/working-with-loras) for the FLUX.1-dev text-to-image model. It can be used with diffusers or ComfyUI.
It was trained on [Replicate](https://replicate.com/) using AI toolkit: https://replicate.com/ostris/flux-dev-lora-trainer/train
## Trigger words
You should use `Sasha` to trigger the image generation.
## Run this LoRA with an API using Replicate
```py
import replicate
input = {
"prompt": "Sasha",
"lora_weights": "https://huggingface.co/gsoler/sashamori/resolve/main/lora.safetensors"
}
output = replicate.run(
"black-forest-labs/flux-dev-lora",
input=input
)
for index, item in enumerate(output):
with open(f"output_{index}.webp", "wb") as file:
file.write(item.read())
```
## Use it with the [🧨 diffusers library](https://github.com/huggingface/diffusers)
```py
from diffusers import AutoPipelineForText2Image
import torch
pipeline = AutoPipelineForText2Image.from_pretrained('black-forest-labs/FLUX.1-dev', torch_dtype=torch.float16).to('cuda')
pipeline.load_lora_weights('gsoler/sashamori', weight_name='lora.safetensors')
image = pipeline('Sasha').images[0]
```
For more details, including weighting, merging and fusing LoRAs, check the [documentation on loading LoRAs in diffusers](https://huggingface.co/docs/diffusers/main/en/using-diffusers/loading_adapters)
## Training details
- Steps: 2000
- Learning rate: 0.0004
- LoRA rank: 16
## Contribute your own examples
You can use the [community tab](https://huggingface.co/gsoler/sashamori/discussions) to add images that show off what you’ve made with this LoRA.
|
Benjaminpwh/xls-r-300m-toratan-120
|
Benjaminpwh
| 2025-04-28T05:51:55Z | 5 | 0 |
transformers
|
[
"transformers",
"tensorboard",
"safetensors",
"wav2vec2",
"automatic-speech-recognition",
"generated_from_trainer",
"base_model:Benjaminpwh/xlsr-toratan-240-copt-base_K",
"base_model:finetune:Benjaminpwh/xlsr-toratan-240-copt-base_K",
"endpoints_compatible",
"region:us"
] |
automatic-speech-recognition
| 2025-03-28T19:51:05Z |
---
library_name: transformers
base_model: Benjaminpwh/xlsr-toratan-240-copt-base_K
tags:
- generated_from_trainer
model-index:
- name: xls-r-300m-toratan-120
results: []
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# xls-r-300m-toratan-120
This model is a fine-tuned version of [Benjaminpwh/xlsr-toratan-240-copt-base_K](https://huggingface.co/Benjaminpwh/xlsr-toratan-240-copt-base_K) on the None dataset.
It achieves the following results on the evaluation set:
- Loss: 0.0121
- Cer: 0.0036
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 0.0003
- train_batch_size: 16
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 32
- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 500
- num_epochs: 30
- mixed_precision_training: Native AMP
### Training results
| Training Loss | Epoch | Step | Validation Loss | Cer |
|:-------------:|:-------:|:----:|:---------------:|:------:|
| 4.4013 | 4.1667 | 400 | 1.6662 | 0.4168 |
| 1.4361 | 8.3333 | 800 | 0.7861 | 0.2370 |
| 0.8924 | 12.5 | 1200 | 0.4475 | 0.1519 |
| 0.575 | 16.6667 | 1600 | 0.2061 | 0.0790 |
| 0.353 | 20.8333 | 2000 | 0.0794 | 0.0311 |
| 0.2164 | 25.0 | 2400 | 0.0308 | 0.0113 |
| 0.1399 | 29.1667 | 2800 | 0.0121 | 0.0036 |
### Framework versions
- Transformers 4.52.0.dev0
- Pytorch 2.6.0+cu124
- Datasets 3.5.0
- Tokenizers 0.21.1
|
notzero/modellv4
|
notzero
| 2025-04-28T05:47:49Z | 0 | 0 |
transformers
|
[
"transformers",
"pytorch",
"qwen2",
"text-generation",
"unsloth",
"trl",
"sft",
"conversational",
"arxiv:1910.09700",
"autotrain_compatible",
"text-generation-inference",
"endpoints_compatible",
"region:us"
] |
text-generation
| 2025-04-27T20:38:28Z |
---
library_name: transformers
tags:
- unsloth
- trl
- sft
---
# Model Card for Model ID
<!-- Provide a quick summary of what the model is/does. -->
## Model Details
### Model Description
<!-- Provide a longer summary of what this model is. -->
This is the model card of a 🤗 transformers model that has been pushed on the Hub. This model card has been automatically generated.
- **Developed by:** [More Information Needed]
- **Funded by [optional]:** [More Information Needed]
- **Shared by [optional]:** [More Information Needed]
- **Model type:** [More Information Needed]
- **Language(s) (NLP):** [More Information Needed]
- **License:** [More Information Needed]
- **Finetuned from model [optional]:** [More Information Needed]
### Model Sources [optional]
<!-- Provide the basic links for the model. -->
- **Repository:** [More Information Needed]
- **Paper [optional]:** [More Information Needed]
- **Demo [optional]:** [More Information Needed]
## Uses
<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
### Direct Use
<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
[More Information Needed]
### Downstream Use [optional]
<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
[More Information Needed]
### Out-of-Scope Use
<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
[More Information Needed]
## Bias, Risks, and Limitations
<!-- This section is meant to convey both technical and sociotechnical limitations. -->
[More Information Needed]
### Recommendations
<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
## How to Get Started with the Model
Use the code below to get started with the model.
[More Information Needed]
## Training Details
### Training Data
<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
[More Information Needed]
### Training Procedure
<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
#### Preprocessing [optional]
[More Information Needed]
#### Training Hyperparameters
- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
#### Speeds, Sizes, Times [optional]
<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
[More Information Needed]
## Evaluation
<!-- This section describes the evaluation protocols and provides the results. -->
### Testing Data, Factors & Metrics
#### Testing Data
<!-- This should link to a Dataset Card if possible. -->
[More Information Needed]
#### Factors
<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
[More Information Needed]
#### Metrics
<!-- These are the evaluation metrics being used, ideally with a description of why. -->
[More Information Needed]
### Results
[More Information Needed]
#### Summary
## Model Examination [optional]
<!-- Relevant interpretability work for the model goes here -->
[More Information Needed]
## Environmental Impact
<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
- **Hardware Type:** [More Information Needed]
- **Hours used:** [More Information Needed]
- **Cloud Provider:** [More Information Needed]
- **Compute Region:** [More Information Needed]
- **Carbon Emitted:** [More Information Needed]
## Technical Specifications [optional]
### Model Architecture and Objective
[More Information Needed]
### Compute Infrastructure
[More Information Needed]
#### Hardware
[More Information Needed]
#### Software
[More Information Needed]
## Citation [optional]
<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
**BibTeX:**
[More Information Needed]
**APA:**
[More Information Needed]
## Glossary [optional]
<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
[More Information Needed]
## More Information [optional]
[More Information Needed]
## Model Card Authors [optional]
[More Information Needed]
## Model Card Contact
[More Information Needed]
|
sLxOpUhh345X/hayay
|
sLxOpUhh345X
| 2025-04-28T05:47:08Z | 0 | 0 | null |
[
"license:bigscience-bloom-rail-1.0",
"region:us"
] | null | 2025-04-28T05:47:08Z |
---
license: bigscience-bloom-rail-1.0
---
|
cata2002/llama-3-8b-full-dataset
|
cata2002
| 2025-04-28T05:45:14Z | 0 | 0 |
transformers
|
[
"transformers",
"gguf",
"llama",
"text-generation-inference",
"unsloth",
"en",
"base_model:unsloth/llama-3-8b-Instruct-bnb-4bit",
"base_model:quantized:unsloth/llama-3-8b-Instruct-bnb-4bit",
"license:apache-2.0",
"endpoints_compatible",
"region:us",
"conversational"
] | null | 2025-04-28T05:43:59Z |
---
base_model: unsloth/llama-3-8b-Instruct-bnb-4bit
tags:
- text-generation-inference
- transformers
- unsloth
- llama
- gguf
license: apache-2.0
language:
- en
---
# Uploaded model
- **Developed by:** cata2002
- **License:** apache-2.0
- **Finetuned from model :** unsloth/llama-3-8b-Instruct-bnb-4bit
This llama model was trained 2x faster with [Unsloth](https://github.com/unslothai/unsloth) and Huggingface's TRL library.
[<img src="https://raw.githubusercontent.com/unslothai/unsloth/main/images/unsloth%20made%20with%20love.png" width="200"/>](https://github.com/unslothai/unsloth)
|
MLconArtist/gemma-3-finetune
|
MLconArtist
| 2025-04-28T05:44:05Z | 0 | 0 |
transformers
|
[
"transformers",
"gemma3_text",
"text-generation",
"text-generation-inference",
"unsloth",
"gemma3",
"conversational",
"en",
"base_model:unsloth/gemma-3-4b-it",
"base_model:finetune:unsloth/gemma-3-4b-it",
"license:apache-2.0",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] |
text-generation
| 2025-04-28T05:43:12Z |
---
base_model: unsloth/gemma-3-4b-it
tags:
- text-generation-inference
- transformers
- unsloth
- gemma3
license: apache-2.0
language:
- en
---
# Uploaded finetuned model
- **Developed by:** MLconArtist
- **License:** apache-2.0
- **Finetuned from model :** unsloth/gemma-3-4b-it
This gemma3 model was trained 2x faster with [Unsloth](https://github.com/unslothai/unsloth) and Huggingface's TRL library.
[<img src="https://raw.githubusercontent.com/unslothai/unsloth/main/images/unsloth%20made%20with%20love.png" width="200"/>](https://github.com/unslothai/unsloth)
|
YOYO-AI/Qwen2.5-32B-YOYO-karcher-base
|
YOYO-AI
| 2025-04-28T05:44:03Z | 0 | 0 |
transformers
|
[
"transformers",
"safetensors",
"qwen2",
"text-generation",
"mergekit",
"merge",
"conversational",
"base_model:Azure99/Blossom-V6-32B",
"base_model:merge:Azure99/Blossom-V6-32B",
"base_model:EVA-UNIT-01/EVA-Qwen2.5-32B-v0.2",
"base_model:merge:EVA-UNIT-01/EVA-Qwen2.5-32B-v0.2",
"base_model:Qwen/Qwen2.5-32B",
"base_model:merge:Qwen/Qwen2.5-32B",
"base_model:arcee-ai/Virtuoso-Medium-v2",
"base_model:merge:arcee-ai/Virtuoso-Medium-v2",
"autotrain_compatible",
"text-generation-inference",
"endpoints_compatible",
"region:us"
] |
text-generation
| 2025-04-28T04:39:15Z |
---
base_model:
- Azure99/Blossom-V6-32B
- arcee-ai/Virtuoso-Medium-v2
- EVA-UNIT-01/EVA-Qwen2.5-32B-v0.2
- Qwen/Qwen2.5-32B
library_name: transformers
tags:
- mergekit
- merge
---
# merge
This is a merge of pre-trained language models created using [mergekit](https://github.com/cg123/mergekit).
## Merge Details
### Merge Method
This model was merged using the [Karcher Mean](https://en.wikipedia.org/wiki/Karcher_mean) merge method using [Qwen/Qwen2.5-32B](https://huggingface.co/Qwen/Qwen2.5-32B) as a base.
### Models Merged
The following models were included in the merge:
* [Azure99/Blossom-V6-32B](https://huggingface.co/Azure99/Blossom-V6-32B)
* [arcee-ai/Virtuoso-Medium-v2](https://huggingface.co/arcee-ai/Virtuoso-Medium-v2)
* [EVA-UNIT-01/EVA-Qwen2.5-32B-v0.2](https://huggingface.co/EVA-UNIT-01/EVA-Qwen2.5-32B-v0.2)
### Configuration
The following YAML configuration was used to produce this model:
```yaml
models:
- model: EVA-UNIT-01/EVA-Qwen2.5-32B-v0.2
- model: Azure99/Blossom-V6-32B
- model: arcee-ai/Virtuoso-Medium-v2
merge_method: karcher
base_model: Qwen/Qwen2.5-32B
parameters:
max_iter: 1000
normalize: true
int8_mask: true
tokenizer_source: base
dtype: float16
```
|
Nourix545/Nourix43434
|
Nourix545
| 2025-04-28T05:43:22Z | 0 | 0 | null |
[
"region:us"
] | null | 2025-04-28T05:40:50Z |
Nourix est un complément alimentaire haut de gamme à base de plantes conçu pour favoriser la gestion naturelle du poids et le bien-être holistique. Conçu pour ceux qui recherchent une approche équilibrée de la santé, il combine des ingrédients scientifiquement prouvés pour stimuler le métabolisme, supprimer l'appétit, augmenter l'énergie et favoriser la détoxification.
## **[Cliquez ici pour commander sur le site officiel de Nourix](https://nourixfrance.fr)**
## Qu'est-ce que Nourix ?
Nourix est un complément alimentaire formulé pour favoriser une gestion saine du poids en ciblant plusieurs aspects du métabolisme, du contrôle de l'appétit et des niveaux d'énergie. Contrairement à de nombreux produits de perte de poids qui reposent sur des stimulants agressifs ou des régimes restrictifs, Nourix adopte une approche holistique et utilise des ingrédients naturels pour soutenir les processus naturels du corps. Il est commercialisé comme un produit végétalien, sans gluten et sans OGM, ce qui le rend adapté à un large éventail de préférences alimentaires.
## Ingrédients clés et leurs bienfaits
La formule de Nourix est construite sur un mélange synergique d'ingrédients naturels, chacun sélectionné pour son rôle dans le soutien de la gestion du poids. Voici un aperçu de certains des principaux composants :
Extrait de thé vert (300 mg) : Le thé vert, riche en catéchines comme l'EGCG, est un stimulant métabolique bien connu. Il favorise la thermogenèse, aidant le corps à brûler les calories plus efficacement, même au repos. Il offre également des bienfaits antioxydants qui favorisent la santé globale.
Berbérine HCL : Ce composé, dérivé de plantes comme l'épine-vinette, aide à réguler la glycémie et favorise la dégradation des graisses. En stabilisant le glucose, il peut réduire les envies et empêcher le stockage des graisses causé par les pics d’insuline.
Gingembre : Traditionnellement utilisé pour ses propriétés thermogéniques, le gingembre améliore la combustion des calories et facilite la digestion, réduisant les ballonnements et améliorant la santé intestinale.
Cannelle : Connue pour stabiliser la glycémie, la cannelle aide à réduire les envies de sucre et favorise le contrôle de l'appétit, ce qui facilite le respect d'une alimentation équilibrée.
Vinaigre de cidre : Cet ingrédient favorise la digestion, régule l'appétit et peut augmenter le métabolisme des graisses, contribuant ainsi à une sensation de satiété.
Piment de Cayenne (Cayenne Pepper) : Thermogénique naturel, le poivre de Cayenne stimule le métabolisme et favorise l'oxydation des graisses, aidant le corps à brûler plus de calories.
Chardon-Marie (Chardon-Marie) : Inclus dans certaines formulations, le chardon-Marie soutient la santé du foie, aide à la détoxification et améliore la capacité du corps à traiter les graisses.
Ces ingrédients agissent ensemble pour accélérer le métabolisme, réduire l’appétit, stabiliser la glycémie et stimuler l’énergie, offrant une approche globale de la gestion du poids sans recourir à des stimulants agressifs.
## Comment fonctionne Nourix ?
**Nourix opère selon une stratégie à plusieurs volets pour soutenir la perte de poids et le bien-être général :**
Stimulation du métabolisme : des ingrédients comme le thé vert, le poivre de Cayenne et le gingembre stimulent la thermogenèse et augmentent la capacité de combustion des calories du corps, même pendant les périodes de repos.
Contrôle de l'appétit : des composants comme la cannelle, le vinaigre de cidre de pomme et la feuille de banaba aident à réguler la glycémie et à favoriser la satiété, réduisant ainsi les fringales et la suralimentation.
Boost d'énergie : Le ginseng, la vitamine B6/B12 et le resvératrol fournissent un regain d'énergie constant qui combat la fatigue sans la nervosité associée aux doses élevées de caféine.
Soutien et détoxification du foie : des ingrédients comme le chardon-Marie et la racine de pissenlit favorisent la santé du foie, aidant le corps à éliminer les toxines et à traiter les graisses plus efficacement.
## **[Cliquez ici pour commander sur le site officiel de Nourix](https://nourixfrance.fr)**
Pour de meilleurs résultats, la dose recommandée est de deux capsules par jour, prises avec un grand verre d'eau, de préférence pendant un repas pour augmenter l'absorption et minimiser les troubles digestifs. Au moins 2 à 3 mois d’utilisation régulière, associés à une alimentation équilibrée et à un exercice modéré, sont recommandés pour des résultats notables.
## Avantages de Nourix
**Nourix offre plusieurs avantages qui en font un choix incontournable pour ceux qui recherchent une solution naturelle de gestion du poids :**
Naturel et sûr : la formule est exempte d’additifs artificiels, d’OGM, de gluten et d’allergènes majeurs, ce qui la rend bien tolérée par la plupart des utilisateurs. Les effets secondaires, tels qu’une légère indigestion, sont rares et disparaissent généralement rapidement.
Approche holistique : En agissant sur le métabolisme, l’appétit, l’énergie et la détoxification, Nourix favorise une perte de poids durable plutôt qu’une perte d’eau temporaire.
Facilité d'utilisation : Les capsules sont faciles à intégrer dans les routines quotidiennes et ne nécessitent aucun rituel complexe.
Commentaires positifs des utilisateurs : De nombreux utilisateurs signalent une réduction des envies, une augmentation de l'énergie et une perte de poids progressive (5 à 7 kg sur 1 à 2 mois) lorsqu'elles sont associées à un mode de vie sain. Les avis soulignent l’amélioration de la digestion et de la clarté mentale comme avantages supplémentaires.
Garantie de remboursement : Le fabricant offre une garantie de satisfaction de 30 jours, ce qui permet aux utilisateurs d'essayer Nourix sans risque.
## Nourix est-il légitime ?
La légitimité de Nourix est mitigée. D’une part, sa formulation est basée sur des ingrédients bien étudiés et les avis positifs des utilisateurs suggèrent qu’elle peut être efficace lorsqu’elle est utilisée dans le cadre d’un mode de vie sain. La conformité du produit aux normes HACCP et l'approbation de la FDA pour la qualité (comme le prétendent certaines sources) renforcent encore sa crédibilité.
En revanche, les critiques négatives et les avertissements concernant des sites Web peu fiables sont des signaux d’alarme. Le manque de transparence sur certains sites Web affiliés à Nourix et les rapports de frais non autorisés suggèrent que les consommateurs devraient faire preuve de prudence. Pour garantir un achat sûr, achetez uniquement auprès des canaux officiels et consultez un professionnel de la santé avant de commencer tout supplément, surtout si vous avez des problèmes de santé préexistants ou si vous prenez des médicaments.
## Réflexions finales
Nourix propose une approche prometteuse et naturelle de la gestion du poids qui utilise un mélange d'ingrédients scientifiquement prouvés pour stimuler le métabolisme, contrôler l'appétit et améliorer l'énergie. Sa formule holistique et sa facilité d’utilisation en font une option intéressante pour ceux qui recherchent une perte de poids durable sans mesures extrêmes. Cependant, les acheteurs potentiels doivent se méfier des produits contrefaits et des vendeurs non vérifiés et s'en tenir aux sites Web officiels pour leurs achats.
## **[Cliquez ici pour commander sur le site officiel de Nourix](https://nourixfrance.fr)**
|
nsadeq/ReDis-Llama
|
nsadeq
| 2025-04-28T05:42:59Z | 0 | 0 |
transformers
|
[
"transformers",
"safetensors",
"Inductive",
"Reasoning",
"text-generation",
"en",
"dataset:nsadeq/redis_generate_rule_alignment",
"dataset:nsadeq/redis_generate_rule_sft",
"dataset:nsadeq/redis_follow_rule_sft",
"arxiv:2504.10647",
"base_model:meta-llama/Meta-Llama-3-8B-Instruct",
"base_model:finetune:meta-llama/Meta-Llama-3-8B-Instruct",
"endpoints_compatible",
"region:us"
] |
text-generation
| 2025-04-27T13:51:47Z |
---
library_name: transformers
tags:
- Inductive
- Reasoning
language:
- en
base_model:
- meta-llama/Meta-Llama-3-8B-Instruct
pipeline_tag: text-generation
datasets:
- nsadeq/redis_generate_rule_alignment
- nsadeq/redis_generate_rule_sft
- nsadeq/redis_follow_rule_sft
---
# Model Card for Model ID
ReDis-Llama is trained for improved inductive reasoning performance.
### Model Description
- **Developed by:** Nafis Sadeq
- **Language(s) (NLP):** English
- **Finetuned from model:** meta-llama/Meta-Llama-3-8B-Instruct
### Model Sources [optional]
- **Repository:** https://github.com/NafisSadeq/reasoning-distillation
- **Paper:** https://arxiv.org/abs/2504.10647
## How to Get Started with the Model
Follow the instructions here: https://github.com/NafisSadeq/reasoning-distillation
## Training Details
Training details can be found in the paper: https://arxiv.org/abs/2504.10647
## Environmental Impact
- **Hardware Type:** 2 × 48 GB Nvidia RTX A6000 GPUs
- **Hours used:** 72 hours
### Model Architecture and Objective
This model has the same architecture as meta-llama/Meta-Llama-3-8B-Instruct
### Compute Infrastructure
2 × 48 GB Nvidia RTX A6000 GPUs
## Citation
If you use this model, please cite the following paper.
@misc{sadeq2025improvingincontextlearningreasoning,
title={Improving In-Context Learning with Reasoning Distillation},
author={Nafis Sadeq and Xin Xu and Zhouhang Xie and Julian McAuley and Byungkyu Kang and Prarit Lamba and Xiang Gao},
year={2025},
eprint={2504.10647},
archivePrefix={arXiv},
primaryClass={cs.CL},
url={https://arxiv.org/abs/2504.10647},
}
|
Triangle104/GLM4-9B-Neon-v2-Q6_K-GGUF
|
Triangle104
| 2025-04-28T05:40:37Z | 0 | 0 |
transformers
|
[
"transformers",
"gguf",
"llama-cpp",
"gguf-my-repo",
"en",
"dataset:allura-org/Celeste-Filtered",
"dataset:allura-org/neon-41k",
"dataset:EVA-UNIT-01/Lilith-v0.2",
"base_model:allura-org/GLM4-9B-Neon-v2",
"base_model:quantized:allura-org/GLM4-9B-Neon-v2",
"license:mit",
"endpoints_compatible",
"region:us",
"conversational"
] | null | 2025-04-28T05:32:25Z |
---
base_model: allura-org/GLM4-9B-Neon-v2
datasets:
- allura-org/Celeste-Filtered
- allura-org/neon-41k
- EVA-UNIT-01/Lilith-v0.2
language:
- en
library_name: transformers
license: mit
tags:
- llama-cpp
- gguf-my-repo
---
# Triangle104/GLM4-9B-Neon-v2-Q6_K-GGUF
This model was converted to GGUF format from [`allura-org/GLM4-9B-Neon-v2`](https://huggingface.co/allura-org/GLM4-9B-Neon-v2) using llama.cpp via the ggml.ai's [GGUF-my-repo](https://huggingface.co/spaces/ggml-org/gguf-my-repo) space.
Refer to the [original model card](https://huggingface.co/allura-org/GLM4-9B-Neon-v2) for more details on the model.
---
RP finetune of GLM-4-9B-0414. Feels nice, lots of personality, if bit
quirky sometimes. Nice prose, not too Claude-ish or Gemini-ish. Doesn't
seem to like too long system prompts or charcards though. Seems to like
JSON formatted system prompts.
Model was trained by Auri.
Training notes
-
Model was trained on a dataset consisting of 77M tokens of synthetic
RP and short story gen data for one epoch. Training took around 11 hours
on 2xRTX 3090 workstation, generously provided by OwenArli.
Went with some sane defaults for training config, QLoRA plus CCE for a
nice chunk of memory usage optimization, 16k fit on 48GB nicely with
some room to spare. I seem to have a problem with Eval/Loss being
broken, not sure why, otherwise it trained smoothly.
Huge thanks to ArliAI for providing compute and collaborating on this run!
Format
-
Model responds to GLM4 instruct formatting, exactly like it's base
model. Backends struggle to add BOS token automatically, so you'll need
to do it yourself. Jinja template should work for chat completions.
[gMASK]<sop><|system|>
{system_prompt}<|user|>
{prompt}<|assistant|>
Recommended Samplers
-
Nothing special, just classics.
Temperature - 1
Min-P - 0.1
Repetition Penalty - 1.03
Example master import for SillyTavern (using Shingane-v1 system prompt by Steelskull)
Running on KoboldCPP and other backends
-
To run GGUFs correctly, you need the most recent version of KoboldCPP, and to pass --overridekv glm4.rope.dimension_count=int:64 to the CLI command or put glm4.rope.dimension_count=int:64 into overridekv box in the GUI (under the Tokens tab at the very bottom).
Thanks to DaringDuck and tofumagnate for info how to apply this fix.
To run this model on vLLM, you'll need to build it from source from the git repo, full GLM4 support hasn't reached release yet.
ExLLaMAv2 and v3 based backends, such as TabbyAPI should support the model out of the box.
Latest versions of llama.cpp server should also allow running GGUFs out-of-the-box.
---
## Use with llama.cpp
Install llama.cpp through brew (works on Mac and Linux)
```bash
brew install llama.cpp
```
Invoke the llama.cpp server or the CLI.
### CLI:
```bash
llama-cli --hf-repo Triangle104/GLM4-9B-Neon-v2-Q6_K-GGUF --hf-file glm4-9b-neon-v2-q6_k.gguf -p "The meaning to life and the universe is"
```
### Server:
```bash
llama-server --hf-repo Triangle104/GLM4-9B-Neon-v2-Q6_K-GGUF --hf-file glm4-9b-neon-v2-q6_k.gguf -c 2048
```
Note: You can also use this checkpoint directly through the [usage steps](https://github.com/ggerganov/llama.cpp?tab=readme-ov-file#usage) listed in the Llama.cpp repo as well.
Step 1: Clone llama.cpp from GitHub.
```
git clone https://github.com/ggerganov/llama.cpp
```
Step 2: Move into the llama.cpp folder and build it with `LLAMA_CURL=1` flag along with other hardware-specific flags (for ex: LLAMA_CUDA=1 for Nvidia GPUs on Linux).
```
cd llama.cpp && LLAMA_CURL=1 make
```
Step 3: Run inference through the main binary.
```
./llama-cli --hf-repo Triangle104/GLM4-9B-Neon-v2-Q6_K-GGUF --hf-file glm4-9b-neon-v2-q6_k.gguf -p "The meaning to life and the universe is"
```
or
```
./llama-server --hf-repo Triangle104/GLM4-9B-Neon-v2-Q6_K-GGUF --hf-file glm4-9b-neon-v2-q6_k.gguf -c 2048
```
|
arkitex/wav2vec2-finetune-authentic-and-synth
|
arkitex
| 2025-04-28T05:39:26Z | 5 | 0 |
transformers
|
[
"transformers",
"tensorboard",
"safetensors",
"wav2vec2",
"automatic-speech-recognition",
"generated_from_trainer",
"dataset:common_voice_17_0",
"base_model:arkitex/wav2vec2-finetune-authentic-cv-only-colab",
"base_model:finetune:arkitex/wav2vec2-finetune-authentic-cv-only-colab",
"license:apache-2.0",
"model-index",
"endpoints_compatible",
"region:us"
] |
automatic-speech-recognition
| 2025-04-24T21:46:57Z |
---
library_name: transformers
license: apache-2.0
base_model: arkitex/wav2vec2-finetune-authentic-cv-only-colab
tags:
- generated_from_trainer
datasets:
- common_voice_17_0
metrics:
- wer
model-index:
- name: wav2vec2-finetune-authentic-and-synth
results:
- task:
name: Automatic Speech Recognition
type: automatic-speech-recognition
dataset:
name: common_voice_17_0
type: common_voice_17_0
config: en
split: None
args: en
metrics:
- name: Wer
type: wer
value: 0.37995702285602656
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# wav2vec2-finetune-authentic-and-synth
This model is a fine-tuned version of [arkitex/wav2vec2-finetune-authentic-cv-only-colab](https://huggingface.co/arkitex/wav2vec2-finetune-authentic-cv-only-colab) on the common_voice_17_0 dataset.
It achieves the following results on the evaluation set:
- Loss: 0.8896
- Wer: 0.3800
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 5e-06
- train_batch_size: 2
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 4
- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 500
- num_epochs: 5
### Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|:-------------:|:------:|:----:|:---------------:|:------:|
| 1.0356 | 0.3761 | 500 | 0.9584 | 0.3776 |
| 1.0564 | 0.7522 | 1000 | 0.9473 | 0.3753 |
| 0.9867 | 1.1279 | 1500 | 0.9273 | 0.3778 |
| 0.9706 | 1.5039 | 2000 | 0.9131 | 0.3792 |
| 0.9542 | 1.8800 | 2500 | 0.9029 | 0.3805 |
| 0.9127 | 2.2557 | 3000 | 0.8947 | 0.3805 |
| 0.9099 | 2.6318 | 3500 | 0.8928 | 0.3813 |
| 0.9128 | 3.0075 | 4000 | 0.8898 | 0.3802 |
| 0.9138 | 3.3836 | 4500 | 0.8907 | 0.3802 |
| 0.8905 | 3.7597 | 5000 | 0.8896 | 0.3800 |
### Framework versions
- Transformers 4.51.3
- Pytorch 2.6.0
- Datasets 3.5.0
- Tokenizers 0.21.0
|
Triangle104/Qwen2.5-0.5B-Q4_K_S-GGUF
|
Triangle104
| 2025-04-28T05:36:47Z | 4 | 0 | null |
[
"gguf",
"llama-cpp",
"gguf-my-repo",
"text-generation",
"zho",
"eng",
"fra",
"spa",
"por",
"deu",
"ita",
"rus",
"jpn",
"kor",
"vie",
"tha",
"ara",
"base_model:Qwen/Qwen2.5-0.5B",
"base_model:quantized:Qwen/Qwen2.5-0.5B",
"license:apache-2.0",
"endpoints_compatible",
"region:us",
"conversational"
] |
text-generation
| 2024-09-22T17:53:22Z |
---
base_model: Qwen/Qwen2.5-0.5B
language:
- zho
- eng
- fra
- spa
- por
- deu
- ita
- rus
- jpn
- kor
- vie
- tha
- ara
license: apache-2.0
license_link: https://huggingface.co/Qwen/Qwen2.5-0.5B/blob/main/LICENSE
pipeline_tag: text-generation
tags:
- llama-cpp
- gguf-my-repo
---
# Triangle104/Qwen2.5-0.5B-Q4_K_S-GGUF
This model was converted to GGUF format from [`Qwen/Qwen2.5-0.5B`](https://huggingface.co/Qwen/Qwen2.5-0.5B) using llama.cpp via the ggml.ai's [GGUF-my-repo](https://huggingface.co/spaces/ggml-org/gguf-my-repo) space.
Refer to the [original model card](https://huggingface.co/Qwen/Qwen2.5-0.5B) for more details on the model.
## Use with llama.cpp
Install llama.cpp through brew (works on Mac and Linux)
```bash
brew install llama.cpp
```
Invoke the llama.cpp server or the CLI.
### CLI:
```bash
llama-cli --hf-repo Triangle104/Qwen2.5-0.5B-Q4_K_S-GGUF --hf-file qwen2.5-0.5b-q4_k_s.gguf -p "The meaning to life and the universe is"
```
### Server:
```bash
llama-server --hf-repo Triangle104/Qwen2.5-0.5B-Q4_K_S-GGUF --hf-file qwen2.5-0.5b-q4_k_s.gguf -c 2048
```
Note: You can also use this checkpoint directly through the [usage steps](https://github.com/ggerganov/llama.cpp?tab=readme-ov-file#usage) listed in the Llama.cpp repo as well.
Step 1: Clone llama.cpp from GitHub.
```
git clone https://github.com/ggerganov/llama.cpp
```
Step 2: Move into the llama.cpp folder and build it with `LLAMA_CURL=1` flag along with other hardware-specific flags (for ex: LLAMA_CUDA=1 for Nvidia GPUs on Linux).
```
cd llama.cpp && LLAMA_CURL=1 make
```
Step 3: Run inference through the main binary.
```
./llama-cli --hf-repo Triangle104/Qwen2.5-0.5B-Q4_K_S-GGUF --hf-file qwen2.5-0.5b-q4_k_s.gguf -p "The meaning to life and the universe is"
```
or
```
./llama-server --hf-repo Triangle104/Qwen2.5-0.5B-Q4_K_S-GGUF --hf-file qwen2.5-0.5b-q4_k_s.gguf -c 2048
```
|
Triangle104/Qwen2.5-1.5B-Q6_K-GGUF
|
Triangle104
| 2025-04-28T05:35:02Z | 4 | 0 | null |
[
"gguf",
"llama-cpp",
"gguf-my-repo",
"text-generation",
"zho",
"eng",
"fra",
"spa",
"por",
"deu",
"ita",
"rus",
"jpn",
"kor",
"vie",
"tha",
"ara",
"base_model:Qwen/Qwen2.5-1.5B",
"base_model:quantized:Qwen/Qwen2.5-1.5B",
"license:apache-2.0",
"endpoints_compatible",
"region:us",
"conversational"
] |
text-generation
| 2024-09-22T17:34:11Z |
---
base_model: Qwen/Qwen2.5-1.5B
language:
- zho
- eng
- fra
- spa
- por
- deu
- ita
- rus
- jpn
- kor
- vie
- tha
- ara
license: apache-2.0
license_link: https://huggingface.co/Qwen/Qwen2.5-1.5B/blob/main/LICENSE
pipeline_tag: text-generation
tags:
- llama-cpp
- gguf-my-repo
---
# Triangle104/Qwen2.5-1.5B-Q6_K-GGUF
This model was converted to GGUF format from [`Qwen/Qwen2.5-1.5B`](https://huggingface.co/Qwen/Qwen2.5-1.5B) using llama.cpp via the ggml.ai's [GGUF-my-repo](https://huggingface.co/spaces/ggml-org/gguf-my-repo) space.
Refer to the [original model card](https://huggingface.co/Qwen/Qwen2.5-1.5B) for more details on the model.
## Use with llama.cpp
Install llama.cpp through brew (works on Mac and Linux)
```bash
brew install llama.cpp
```
Invoke the llama.cpp server or the CLI.
### CLI:
```bash
llama-cli --hf-repo Triangle104/Qwen2.5-1.5B-Q6_K-GGUF --hf-file qwen2.5-1.5b-q6_k.gguf -p "The meaning to life and the universe is"
```
### Server:
```bash
llama-server --hf-repo Triangle104/Qwen2.5-1.5B-Q6_K-GGUF --hf-file qwen2.5-1.5b-q6_k.gguf -c 2048
```
Note: You can also use this checkpoint directly through the [usage steps](https://github.com/ggerganov/llama.cpp?tab=readme-ov-file#usage) listed in the Llama.cpp repo as well.
Step 1: Clone llama.cpp from GitHub.
```
git clone https://github.com/ggerganov/llama.cpp
```
Step 2: Move into the llama.cpp folder and build it with `LLAMA_CURL=1` flag along with other hardware-specific flags (for ex: LLAMA_CUDA=1 for Nvidia GPUs on Linux).
```
cd llama.cpp && LLAMA_CURL=1 make
```
Step 3: Run inference through the main binary.
```
./llama-cli --hf-repo Triangle104/Qwen2.5-1.5B-Q6_K-GGUF --hf-file qwen2.5-1.5b-q6_k.gguf -p "The meaning to life and the universe is"
```
or
```
./llama-server --hf-repo Triangle104/Qwen2.5-1.5B-Q6_K-GGUF --hf-file qwen2.5-1.5b-q6_k.gguf -c 2048
```
|
Alcoft/Qwen2.5-7B-Instruct-GGUF
|
Alcoft
| 2025-04-28T05:34:48Z | 22 | 0 | null |
[
"gguf",
"text-generation",
"zho",
"eng",
"fra",
"spa",
"por",
"deu",
"ita",
"rus",
"jpn",
"kor",
"vie",
"tha",
"ara",
"base_model:Qwen/Qwen2.5-7B-Instruct",
"base_model:quantized:Qwen/Qwen2.5-7B-Instruct",
"license:apache-2.0",
"endpoints_compatible",
"region:us",
"conversational"
] |
text-generation
| 2024-12-01T01:08:44Z |
---
license: apache-2.0
language:
- zho
- eng
- fra
- spa
- por
- deu
- ita
- rus
- jpn
- kor
- vie
- tha
- ara
base_model:
- Qwen/Qwen2.5-7B-Instruct
pipeline_tag: text-generation
---
|
Triangle104/Qwen2.5-3B-Q4_K_M-GGUF
|
Triangle104
| 2025-04-28T05:34:38Z | 2 | 0 | null |
[
"gguf",
"llama-cpp",
"gguf-my-repo",
"text-generation",
"zho",
"eng",
"fra",
"spa",
"por",
"deu",
"ita",
"rus",
"jpn",
"kor",
"vie",
"tha",
"ara",
"base_model:Qwen/Qwen2.5-3B",
"base_model:quantized:Qwen/Qwen2.5-3B",
"license:other",
"endpoints_compatible",
"region:us",
"conversational"
] |
text-generation
| 2024-09-22T16:49:28Z |
---
base_model: Qwen/Qwen2.5-3B
language:
- zho
- eng
- fra
- spa
- por
- deu
- ita
- rus
- jpn
- kor
- vie
- tha
- ara
license: other
license_name: qwen-research
license_link: https://huggingface.co/Qwen/Qwen2.5-3B/blob/main/LICENSE
pipeline_tag: text-generation
tags:
- llama-cpp
- gguf-my-repo
---
# Triangle104/Qwen2.5-3B-Q4_K_M-GGUF
This model was converted to GGUF format from [`Qwen/Qwen2.5-3B`](https://huggingface.co/Qwen/Qwen2.5-3B) using llama.cpp via the ggml.ai's [GGUF-my-repo](https://huggingface.co/spaces/ggml-org/gguf-my-repo) space.
Refer to the [original model card](https://huggingface.co/Qwen/Qwen2.5-3B) for more details on the model.
## Use with llama.cpp
Install llama.cpp through brew (works on Mac and Linux)
```bash
brew install llama.cpp
```
Invoke the llama.cpp server or the CLI.
### CLI:
```bash
llama-cli --hf-repo Triangle104/Qwen2.5-3B-Q4_K_M-GGUF --hf-file qwen2.5-3b-q4_k_m.gguf -p "The meaning to life and the universe is"
```
### Server:
```bash
llama-server --hf-repo Triangle104/Qwen2.5-3B-Q4_K_M-GGUF --hf-file qwen2.5-3b-q4_k_m.gguf -c 2048
```
Note: You can also use this checkpoint directly through the [usage steps](https://github.com/ggerganov/llama.cpp?tab=readme-ov-file#usage) listed in the Llama.cpp repo as well.
Step 1: Clone llama.cpp from GitHub.
```
git clone https://github.com/ggerganov/llama.cpp
```
Step 2: Move into the llama.cpp folder and build it with `LLAMA_CURL=1` flag along with other hardware-specific flags (for ex: LLAMA_CUDA=1 for Nvidia GPUs on Linux).
```
cd llama.cpp && LLAMA_CURL=1 make
```
Step 3: Run inference through the main binary.
```
./llama-cli --hf-repo Triangle104/Qwen2.5-3B-Q4_K_M-GGUF --hf-file qwen2.5-3b-q4_k_m.gguf -p "The meaning to life and the universe is"
```
or
```
./llama-server --hf-repo Triangle104/Qwen2.5-3B-Q4_K_M-GGUF --hf-file qwen2.5-3b-q4_k_m.gguf -c 2048
```
|
Triangle104/Qwen2.5-3B-Q5_K_M-GGUF
|
Triangle104
| 2025-04-28T05:34:19Z | 2 | 0 | null |
[
"gguf",
"llama-cpp",
"gguf-my-repo",
"text-generation",
"zho",
"eng",
"fra",
"spa",
"por",
"deu",
"ita",
"rus",
"jpn",
"kor",
"vie",
"tha",
"ara",
"base_model:Qwen/Qwen2.5-3B",
"base_model:quantized:Qwen/Qwen2.5-3B",
"license:other",
"endpoints_compatible",
"region:us",
"conversational"
] |
text-generation
| 2024-09-22T16:56:57Z |
---
base_model: Qwen/Qwen2.5-3B
language:
- zho
- eng
- fra
- spa
- por
- deu
- ita
- rus
- jpn
- kor
- vie
- tha
- ara
license: other
license_name: qwen-research
license_link: https://huggingface.co/Qwen/Qwen2.5-3B/blob/main/LICENSE
pipeline_tag: text-generation
tags:
- llama-cpp
- gguf-my-repo
---
# Triangle104/Qwen2.5-3B-Q5_K_M-GGUF
This model was converted to GGUF format from [`Qwen/Qwen2.5-3B`](https://huggingface.co/Qwen/Qwen2.5-3B) using llama.cpp via the ggml.ai's [GGUF-my-repo](https://huggingface.co/spaces/ggml-org/gguf-my-repo) space.
Refer to the [original model card](https://huggingface.co/Qwen/Qwen2.5-3B) for more details on the model.
## Use with llama.cpp
Install llama.cpp through brew (works on Mac and Linux)
```bash
brew install llama.cpp
```
Invoke the llama.cpp server or the CLI.
### CLI:
```bash
llama-cli --hf-repo Triangle104/Qwen2.5-3B-Q5_K_M-GGUF --hf-file qwen2.5-3b-q5_k_m.gguf -p "The meaning to life and the universe is"
```
### Server:
```bash
llama-server --hf-repo Triangle104/Qwen2.5-3B-Q5_K_M-GGUF --hf-file qwen2.5-3b-q5_k_m.gguf -c 2048
```
Note: You can also use this checkpoint directly through the [usage steps](https://github.com/ggerganov/llama.cpp?tab=readme-ov-file#usage) listed in the Llama.cpp repo as well.
Step 1: Clone llama.cpp from GitHub.
```
git clone https://github.com/ggerganov/llama.cpp
```
Step 2: Move into the llama.cpp folder and build it with `LLAMA_CURL=1` flag along with other hardware-specific flags (for ex: LLAMA_CUDA=1 for Nvidia GPUs on Linux).
```
cd llama.cpp && LLAMA_CURL=1 make
```
Step 3: Run inference through the main binary.
```
./llama-cli --hf-repo Triangle104/Qwen2.5-3B-Q5_K_M-GGUF --hf-file qwen2.5-3b-q5_k_m.gguf -p "The meaning to life and the universe is"
```
or
```
./llama-server --hf-repo Triangle104/Qwen2.5-3B-Q5_K_M-GGUF --hf-file qwen2.5-3b-q5_k_m.gguf -c 2048
```
|
Triangle104/Qwen2.5-7B-Q5_K_S-GGUF
|
Triangle104
| 2025-04-28T05:32:53Z | 3 | 0 | null |
[
"gguf",
"llama-cpp",
"gguf-my-repo",
"text-generation",
"zho",
"eng",
"fra",
"spa",
"por",
"deu",
"ita",
"rus",
"jpn",
"kor",
"vie",
"tha",
"ara",
"base_model:Qwen/Qwen2.5-7B",
"base_model:quantized:Qwen/Qwen2.5-7B",
"license:apache-2.0",
"endpoints_compatible",
"region:us",
"conversational"
] |
text-generation
| 2024-09-19T16:05:08Z |
---
base_model: Qwen/Qwen2.5-7B
language:
- zho
- eng
- fra
- spa
- por
- deu
- ita
- rus
- jpn
- kor
- vie
- tha
- ara
license: apache-2.0
license_link: https://huggingface.co/Qwen/Qwen2.5-7B/blob/main/LICENSE
pipeline_tag: text-generation
tags:
- llama-cpp
- gguf-my-repo
---
# Triangle104/Qwen2.5-7B-Q5_K_S-GGUF
This model was converted to GGUF format from [`Qwen/Qwen2.5-7B`](https://huggingface.co/Qwen/Qwen2.5-7B) using llama.cpp via the ggml.ai's [GGUF-my-repo](https://huggingface.co/spaces/ggml-org/gguf-my-repo) space.
Refer to the [original model card](https://huggingface.co/Qwen/Qwen2.5-7B) for more details on the model.
## Use with llama.cpp
Install llama.cpp through brew (works on Mac and Linux)
```bash
brew install llama.cpp
```
Invoke the llama.cpp server or the CLI.
### CLI:
```bash
llama-cli --hf-repo Triangle104/Qwen2.5-7B-Q5_K_S-GGUF --hf-file qwen2.5-7b-q5_k_s.gguf -p "The meaning to life and the universe is"
```
### Server:
```bash
llama-server --hf-repo Triangle104/Qwen2.5-7B-Q5_K_S-GGUF --hf-file qwen2.5-7b-q5_k_s.gguf -c 2048
```
Note: You can also use this checkpoint directly through the [usage steps](https://github.com/ggerganov/llama.cpp?tab=readme-ov-file#usage) listed in the Llama.cpp repo as well.
Step 1: Clone llama.cpp from GitHub.
```
git clone https://github.com/ggerganov/llama.cpp
```
Step 2: Move into the llama.cpp folder and build it with `LLAMA_CURL=1` flag along with other hardware-specific flags (for ex: LLAMA_CUDA=1 for Nvidia GPUs on Linux).
```
cd llama.cpp && LLAMA_CURL=1 make
```
Step 3: Run inference through the main binary.
```
./llama-cli --hf-repo Triangle104/Qwen2.5-7B-Q5_K_S-GGUF --hf-file qwen2.5-7b-q5_k_s.gguf -p "The meaning to life and the universe is"
```
or
```
./llama-server --hf-repo Triangle104/Qwen2.5-7B-Q5_K_S-GGUF --hf-file qwen2.5-7b-q5_k_s.gguf -c 2048
```
|
Triangle104/Qwen2.5-7B-Q5_K_M-GGUF
|
Triangle104
| 2025-04-28T05:32:45Z | 2 | 0 | null |
[
"gguf",
"llama-cpp",
"gguf-my-repo",
"text-generation",
"zho",
"eng",
"fra",
"spa",
"por",
"deu",
"ita",
"rus",
"jpn",
"kor",
"vie",
"tha",
"ara",
"base_model:Qwen/Qwen2.5-7B",
"base_model:quantized:Qwen/Qwen2.5-7B",
"license:apache-2.0",
"endpoints_compatible",
"region:us",
"conversational"
] |
text-generation
| 2024-09-19T16:07:45Z |
---
base_model: Qwen/Qwen2.5-7B
language:
- zho
- eng
- fra
- spa
- por
- deu
- ita
- rus
- jpn
- kor
- vie
- tha
- ara
license: apache-2.0
license_link: https://huggingface.co/Qwen/Qwen2.5-7B/blob/main/LICENSE
pipeline_tag: text-generation
tags:
- llama-cpp
- gguf-my-repo
---
# Triangle104/Qwen2.5-7B-Q5_K_M-GGUF
This model was converted to GGUF format from [`Qwen/Qwen2.5-7B`](https://huggingface.co/Qwen/Qwen2.5-7B) using llama.cpp via the ggml.ai's [GGUF-my-repo](https://huggingface.co/spaces/ggml-org/gguf-my-repo) space.
Refer to the [original model card](https://huggingface.co/Qwen/Qwen2.5-7B) for more details on the model.
## Use with llama.cpp
Install llama.cpp through brew (works on Mac and Linux)
```bash
brew install llama.cpp
```
Invoke the llama.cpp server or the CLI.
### CLI:
```bash
llama-cli --hf-repo Triangle104/Qwen2.5-7B-Q5_K_M-GGUF --hf-file qwen2.5-7b-q5_k_m.gguf -p "The meaning to life and the universe is"
```
### Server:
```bash
llama-server --hf-repo Triangle104/Qwen2.5-7B-Q5_K_M-GGUF --hf-file qwen2.5-7b-q5_k_m.gguf -c 2048
```
Note: You can also use this checkpoint directly through the [usage steps](https://github.com/ggerganov/llama.cpp?tab=readme-ov-file#usage) listed in the Llama.cpp repo as well.
Step 1: Clone llama.cpp from GitHub.
```
git clone https://github.com/ggerganov/llama.cpp
```
Step 2: Move into the llama.cpp folder and build it with `LLAMA_CURL=1` flag along with other hardware-specific flags (for ex: LLAMA_CUDA=1 for Nvidia GPUs on Linux).
```
cd llama.cpp && LLAMA_CURL=1 make
```
Step 3: Run inference through the main binary.
```
./llama-cli --hf-repo Triangle104/Qwen2.5-7B-Q5_K_M-GGUF --hf-file qwen2.5-7b-q5_k_m.gguf -p "The meaning to life and the universe is"
```
or
```
./llama-server --hf-repo Triangle104/Qwen2.5-7B-Q5_K_M-GGUF --hf-file qwen2.5-7b-q5_k_m.gguf -c 2048
```
|
Triangle104/Qwen2.5-14B-Q6_K-GGUF
|
Triangle104
| 2025-04-28T05:31:50Z | 10 | 0 | null |
[
"gguf",
"llama-cpp",
"gguf-my-repo",
"text-generation",
"zho",
"eng",
"fra",
"spa",
"por",
"deu",
"ita",
"rus",
"jpn",
"kor",
"vie",
"tha",
"ara",
"base_model:Qwen/Qwen2.5-14B",
"base_model:quantized:Qwen/Qwen2.5-14B",
"license:apache-2.0",
"endpoints_compatible",
"region:us",
"conversational"
] |
text-generation
| 2024-09-19T14:58:15Z |
---
base_model: Qwen/Qwen2.5-14B
language:
- zho
- eng
- fra
- spa
- por
- deu
- ita
- rus
- jpn
- kor
- vie
- tha
- ara
license: apache-2.0
license_link: https://huggingface.co/Qwen/Qwen2.5-14B/blob/main/LICENSE
pipeline_tag: text-generation
tags:
- llama-cpp
- gguf-my-repo
---
# Triangle104/Qwen2.5-14B-Q6_K-GGUF
This model was converted to GGUF format from [`Qwen/Qwen2.5-14B`](https://huggingface.co/Qwen/Qwen2.5-14B) using llama.cpp via the ggml.ai's [GGUF-my-repo](https://huggingface.co/spaces/ggml-org/gguf-my-repo) space.
Refer to the [original model card](https://huggingface.co/Qwen/Qwen2.5-14B) for more details on the model.
## Use with llama.cpp
Install llama.cpp through brew (works on Mac and Linux)
```bash
brew install llama.cpp
```
Invoke the llama.cpp server or the CLI.
### CLI:
```bash
llama-cli --hf-repo Triangle104/Qwen2.5-14B-Q6_K-GGUF --hf-file qwen2.5-14b-q6_k.gguf -p "The meaning to life and the universe is"
```
### Server:
```bash
llama-server --hf-repo Triangle104/Qwen2.5-14B-Q6_K-GGUF --hf-file qwen2.5-14b-q6_k.gguf -c 2048
```
Note: You can also use this checkpoint directly through the [usage steps](https://github.com/ggerganov/llama.cpp?tab=readme-ov-file#usage) listed in the Llama.cpp repo as well.
Step 1: Clone llama.cpp from GitHub.
```
git clone https://github.com/ggerganov/llama.cpp
```
Step 2: Move into the llama.cpp folder and build it with `LLAMA_CURL=1` flag along with other hardware-specific flags (for ex: LLAMA_CUDA=1 for Nvidia GPUs on Linux).
```
cd llama.cpp && LLAMA_CURL=1 make
```
Step 3: Run inference through the main binary.
```
./llama-cli --hf-repo Triangle104/Qwen2.5-14B-Q6_K-GGUF --hf-file qwen2.5-14b-q6_k.gguf -p "The meaning to life and the universe is"
```
or
```
./llama-server --hf-repo Triangle104/Qwen2.5-14B-Q6_K-GGUF --hf-file qwen2.5-14b-q6_k.gguf -c 2048
```
|
TOMFORD79/S8
|
TOMFORD79
| 2025-04-28T05:27:27Z | 0 | 0 | null |
[
"safetensors",
"any-to-any",
"omega",
"omegalabs",
"bittensor",
"agi",
"license:mit",
"region:us"
] |
any-to-any
| 2025-04-28T04:02:49Z |
---
license: mit
tags:
- any-to-any
- omega
- omegalabs
- bittensor
- agi
---
This is an Any-to-Any model checkpoint for the OMEGA Labs x Bittensor Any-to-Any subnet.
Check out the [git repo](https://github.com/omegalabsinc/omegalabs-anytoany-bittensor) and find OMEGA on X: [@omegalabsai](https://x.com/omegalabsai).
|
Triangle104/Qwen2.5-7B-Instruct-Q5_K_S-GGUF
|
Triangle104
| 2025-04-28T05:25:42Z | 3 | 0 | null |
[
"gguf",
"chat",
"llama-cpp",
"gguf-my-repo",
"text-generation",
"zho",
"eng",
"fra",
"spa",
"por",
"deu",
"ita",
"rus",
"jpn",
"kor",
"vie",
"tha",
"ara",
"base_model:Qwen/Qwen2.5-7B-Instruct",
"base_model:quantized:Qwen/Qwen2.5-7B-Instruct",
"license:apache-2.0",
"endpoints_compatible",
"region:us",
"conversational"
] |
text-generation
| 2024-09-19T15:25:09Z |
---
base_model: Qwen/Qwen2.5-7B-Instruct
language:
- zho
- eng
- fra
- spa
- por
- deu
- ita
- rus
- jpn
- kor
- vie
- tha
- ara
license: apache-2.0
license_link: https://huggingface.co/Qwen/Qwen2.5-7B-Instruct/blob/main/LICENSE
pipeline_tag: text-generation
tags:
- chat
- llama-cpp
- gguf-my-repo
---
# Triangle104/Qwen2.5-7B-Instruct-Q5_K_S-GGUF
This model was converted to GGUF format from [`Qwen/Qwen2.5-7B-Instruct`](https://huggingface.co/Qwen/Qwen2.5-7B-Instruct) using llama.cpp via the ggml.ai's [GGUF-my-repo](https://huggingface.co/spaces/ggml-org/gguf-my-repo) space.
Refer to the [original model card](https://huggingface.co/Qwen/Qwen2.5-7B-Instruct) for more details on the model.
## Use with llama.cpp
Install llama.cpp through brew (works on Mac and Linux)
```bash
brew install llama.cpp
```
Invoke the llama.cpp server or the CLI.
### CLI:
```bash
llama-cli --hf-repo Triangle104/Qwen2.5-7B-Instruct-Q5_K_S-GGUF --hf-file qwen2.5-7b-instruct-q5_k_s.gguf -p "The meaning to life and the universe is"
```
### Server:
```bash
llama-server --hf-repo Triangle104/Qwen2.5-7B-Instruct-Q5_K_S-GGUF --hf-file qwen2.5-7b-instruct-q5_k_s.gguf -c 2048
```
Note: You can also use this checkpoint directly through the [usage steps](https://github.com/ggerganov/llama.cpp?tab=readme-ov-file#usage) listed in the Llama.cpp repo as well.
Step 1: Clone llama.cpp from GitHub.
```
git clone https://github.com/ggerganov/llama.cpp
```
Step 2: Move into the llama.cpp folder and build it with `LLAMA_CURL=1` flag along with other hardware-specific flags (for ex: LLAMA_CUDA=1 for Nvidia GPUs on Linux).
```
cd llama.cpp && LLAMA_CURL=1 make
```
Step 3: Run inference through the main binary.
```
./llama-cli --hf-repo Triangle104/Qwen2.5-7B-Instruct-Q5_K_S-GGUF --hf-file qwen2.5-7b-instruct-q5_k_s.gguf -p "The meaning to life and the universe is"
```
or
```
./llama-server --hf-repo Triangle104/Qwen2.5-7B-Instruct-Q5_K_S-GGUF --hf-file qwen2.5-7b-instruct-q5_k_s.gguf -c 2048
```
|
Yy245/Cat-Advisor
|
Yy245
| 2025-04-28T05:25:20Z | 0 | 0 | null |
[
"gguf",
"license:apache-2.0",
"endpoints_compatible",
"region:us",
"conversational"
] | null | 2025-04-28T03:18:36Z |
---
license: apache-2.0
---
|
archiii/detr-resnet-50-dc5-finetuned
|
archiii
| 2025-04-28T05:24:05Z | 2 | 0 |
transformers
|
[
"transformers",
"safetensors",
"detr",
"object-detection",
"generated_from_trainer",
"base_model:facebook/detr-resnet-50-dc5",
"base_model:finetune:facebook/detr-resnet-50-dc5",
"license:apache-2.0",
"endpoints_compatible",
"region:us"
] |
object-detection
| 2025-04-25T07:32:11Z |
---
library_name: transformers
license: apache-2.0
base_model: facebook/detr-resnet-50-dc5
tags:
- generated_from_trainer
model-index:
- name: detr-resnet-50-dc5-finetuned
results: []
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# detr-resnet-50-dc5-finetuned
This model is a fine-tuned version of [facebook/detr-resnet-50-dc5](https://huggingface.co/facebook/detr-resnet-50-dc5) on the None dataset.
It achieves the following results on the evaluation set:
- Loss: 0.3703
- Map: 0.3703
- Map 50: 0.4284
- Map 75: 0.3912
- Map Small: -1.0
- Map Medium: -1.0
- Map Large: 0.3703
- Mar 1: 0.4491
- Mar 10: 0.5038
- Mar 100: 0.5934
- Mar Small: -1.0
- Mar Medium: -1.0
- Mar Large: 0.5934
- Map Metal: 0.4782
- Mar 100 Metal: 0.6095
- Map Paper: 0.0261
- Mar 100 Paper: 0.4483
- Map Plastic: 0.6064
- Mar 100 Plastic: 0.7225
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 1e-05
- train_batch_size: 4
- eval_batch_size: 4
- seed: 42
- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- training_steps: 10000
- mixed_precision_training: Native AMP
### Training results
| Training Loss | Epoch | Step | Validation Loss | Map | Map 50 | Map 75 | Map Small | Map Medium | Map Large | Mar 1 | Mar 10 | Mar 100 | Mar Small | Mar Medium | Mar Large | Map Background | Mar 100 Background | Map Metal | Mar 100 Metal | Map Paper | Mar 100 Paper | Map Plastic | Mar 100 Plastic |
|:-------------:|:-------:|:-----:|:---------------:|:------:|:------:|:------:|:---------:|:----------:|:---------:|:------:|:------:|:-------:|:---------:|:----------:|:---------:|:--------------:|:------------------:|:---------:|:-------------:|:---------:|:-------------:|:-----------:|:---------------:|
| 1.9796 | 0.1018 | 50 | 2.1283 | 0.0237 | 0.0421 | 0.0238 | -1.0 | -1.0 | 0.0284 | 0.1464 | 0.2542 | 0.3226 | -1.0 | -1.0 | 0.3226 | -1.0 | -1.0 | 0.0546 | 0.6222 | 0.0015 | 0.1103 | 0.015 | 0.2353 |
| 1.9882 | 0.2037 | 100 | 1.6940 | 0.0243 | 0.0456 | 0.0214 | -1.0 | -1.0 | 0.0261 | 0.162 | 0.243 | 0.3188 | -1.0 | -1.0 | 0.3188 | -1.0 | -1.0 | 0.0461 | 0.7143 | 0.0028 | 0.0655 | 0.024 | 0.1765 |
| 2.1489 | 0.3055 | 150 | 1.3699 | 0.0273 | 0.053 | 0.026 | -1.0 | -1.0 | 0.0284 | 0.2013 | 0.2889 | 0.3846 | -1.0 | -1.0 | 0.3846 | -1.0 | -1.0 | 0.0474 | 0.719 | 0.0193 | 0.2759 | 0.0153 | 0.1588 |
| 1.4834 | 0.4073 | 200 | 1.1051 | 0.026 | 0.0432 | 0.0272 | -1.0 | -1.0 | 0.0274 | 0.1959 | 0.2952 | 0.3738 | -1.0 | -1.0 | 0.3738 | -1.0 | -1.0 | 0.06 | 0.846 | 0.0009 | 0.0655 | 0.0172 | 0.2098 |
| 0.8501 | 0.5092 | 250 | 0.9289 | 0.0354 | 0.0526 | 0.0387 | -1.0 | -1.0 | 0.0361 | 0.2408 | 0.3419 | 0.4098 | -1.0 | -1.0 | 0.4098 | -1.0 | -1.0 | 0.0755 | 0.8651 | 0.0017 | 0.1034 | 0.0289 | 0.2608 |
| 1.0472 | 0.6110 | 300 | 0.8789 | 0.0271 | 0.0397 | 0.028 | -1.0 | -1.0 | 0.028 | 0.2794 | 0.4266 | 0.4883 | -1.0 | -1.0 | 0.4883 | -1.0 | -1.0 | 0.0637 | 0.8508 | 0.0027 | 0.1897 | 0.015 | 0.4245 |
| 1.09 | 0.7128 | 350 | 0.8033 | 0.034 | 0.0482 | 0.0364 | -1.0 | -1.0 | 0.0355 | 0.2929 | 0.4048 | 0.4568 | -1.0 | -1.0 | 0.4568 | -1.0 | -1.0 | 0.0848 | 0.8841 | 0.003 | 0.1724 | 0.0142 | 0.3137 |
| 1.1819 | 0.8147 | 400 | 0.7859 | 0.0354 | 0.0485 | 0.039 | -1.0 | -1.0 | 0.0367 | 0.2686 | 0.3644 | 0.4162 | -1.0 | -1.0 | 0.4162 | -1.0 | -1.0 | 0.0947 | 0.8825 | 0.0018 | 0.0966 | 0.0098 | 0.2696 |
| 0.9802 | 0.9165 | 450 | 0.7289 | 0.0409 | 0.0559 | 0.0441 | -1.0 | -1.0 | 0.041 | 0.2321 | 0.3313 | 0.3894 | -1.0 | -1.0 | 0.3894 | -1.0 | -1.0 | 0.1141 | 0.8857 | 0.0008 | 0.0276 | 0.0078 | 0.2549 |
| 0.5236 | 1.0183 | 500 | 0.7572 | 0.0175 | 0.028 | 0.019 | -1.0 | -1.0 | 0.0176 | 0.1883 | 0.3483 | 0.434 | -1.0 | -1.0 | 0.434 | -1.0 | -1.0 | 0.0411 | 0.8254 | 0.0 | 0.0 | 0.0113 | 0.4765 |
| 0.7924 | 1.1202 | 550 | 0.7086 | 0.0306 | 0.0434 | 0.0344 | -1.0 | -1.0 | 0.0309 | 0.2802 | 0.4428 | 0.4935 | -1.0 | -1.0 | 0.4935 | -1.0 | -1.0 | 0.0687 | 0.846 | 0.0001 | 0.0069 | 0.0228 | 0.6275 |
| 1.0112 | 1.2220 | 600 | 0.6612 | 0.0405 | 0.0504 | 0.0436 | -1.0 | -1.0 | 0.0411 | 0.2854 | 0.4361 | 0.4979 | -1.0 | -1.0 | 0.4979 | -1.0 | -1.0 | 0.0938 | 0.8063 | 0.0 | 0.0 | 0.0277 | 0.6873 |
| 0.7764 | 1.3238 | 650 | 0.6513 | 0.031 | 0.0421 | 0.0343 | -1.0 | -1.0 | 0.0331 | 0.2634 | 0.4552 | 0.5265 | -1.0 | -1.0 | 0.5265 | -1.0 | -1.0 | 0.0614 | 0.846 | 0.0013 | 0.0414 | 0.0304 | 0.6922 |
| 0.7316 | 1.4257 | 700 | 0.6150 | 0.0367 | 0.0451 | 0.0409 | -1.0 | -1.0 | 0.0399 | 0.3747 | 0.513 | 0.5556 | -1.0 | -1.0 | 0.5556 | -1.0 | -1.0 | 0.0579 | 0.873 | 0.001 | 0.0448 | 0.0514 | 0.749 |
| 0.5439 | 1.5275 | 750 | 0.7063 | 0.0334 | 0.0471 | 0.038 | -1.0 | -1.0 | 0.0342 | 0.2454 | 0.4007 | 0.4784 | -1.0 | -1.0 | 0.4784 | -1.0 | -1.0 | 0.0568 | 0.6016 | 0.0005 | 0.0621 | 0.0429 | 0.7716 |
| 0.5359 | 1.6293 | 800 | 0.6803 | 0.0407 | 0.0578 | 0.0432 | -1.0 | -1.0 | 0.0415 | 0.2942 | 0.4281 | 0.4878 | -1.0 | -1.0 | 0.4878 | -1.0 | -1.0 | 0.0621 | 0.5841 | 0.0011 | 0.0793 | 0.059 | 0.8 |
| 0.8857 | 1.7312 | 850 | 0.6697 | 0.0539 | 0.077 | 0.0623 | -1.0 | -1.0 | 0.0556 | 0.3079 | 0.431 | 0.507 | -1.0 | -1.0 | 0.507 | -1.0 | -1.0 | 0.0609 | 0.5778 | 0.0353 | 0.1414 | 0.0654 | 0.802 |
| 1.4486 | 1.8330 | 900 | 0.6321 | 0.0436 | 0.0576 | 0.0514 | -1.0 | -1.0 | 0.05 | 0.3419 | 0.473 | 0.5298 | -1.0 | -1.0 | 0.5298 | -1.0 | -1.0 | 0.0473 | 0.6937 | 0.0103 | 0.1034 | 0.0732 | 0.7922 |
| 0.7727 | 1.9348 | 950 | 0.6602 | 0.0298 | 0.0406 | 0.034 | -1.0 | -1.0 | 0.0337 | 0.3176 | 0.4942 | 0.5725 | -1.0 | -1.0 | 0.5725 | -1.0 | -1.0 | 0.0371 | 0.7238 | 0.0108 | 0.2241 | 0.0416 | 0.7696 |
| 1.6886 | 2.0367 | 1000 | 0.6232 | 0.0744 | 0.1046 | 0.0872 | -1.0 | -1.0 | 0.0766 | 0.3938 | 0.5073 | 0.5349 | -1.0 | -1.0 | 0.5349 | -1.0 | -1.0 | 0.15 | 0.7873 | 0.0028 | 0.0241 | 0.0705 | 0.7931 |
| 0.5422 | 2.1385 | 1050 | 0.5752 | 0.0743 | 0.0964 | 0.086 | -1.0 | -1.0 | 0.0809 | 0.4497 | 0.5463 | 0.5649 | -1.0 | -1.0 | 0.5649 | -1.0 | -1.0 | 0.1314 | 0.8349 | 0.008 | 0.0655 | 0.0834 | 0.7941 |
| 0.5052 | 2.2403 | 1100 | 0.5267 | 0.0661 | 0.0825 | 0.0702 | -1.0 | -1.0 | 0.0796 | 0.4281 | 0.5492 | 0.5817 | -1.0 | -1.0 | 0.5817 | -1.0 | -1.0 | 0.0794 | 0.7667 | 0.0424 | 0.0862 | 0.0766 | 0.8922 |
| 0.6956 | 2.3422 | 1150 | 0.5499 | 0.0647 | 0.0912 | 0.0715 | -1.0 | -1.0 | 0.073 | 0.3427 | 0.4879 | 0.5429 | -1.0 | -1.0 | 0.5429 | 0.0743 | 0.6571 | 0.0556 | 0.0724 | 0.0644 | 0.899 |
| 0.5097 | 2.4440 | 1200 | 0.5346 | 0.0748 | 0.0924 | 0.0858 | -1.0 | -1.0 | 0.0819 | 0.4112 | 0.5247 | 0.558 | -1.0 | -1.0 | 0.558 | -1.0 | -1.0 | 0.1015 | 0.7254 | 0.0332 | 0.0379 | 0.0898 | 0.9108 |
| 0.5358 | 2.5458 | 1250 | 0.5622 | 0.095 | 0.1303 | 0.1139 | -1.0 | -1.0 | 0.0971 | 0.4281 | 0.533 | 0.5571 | -1.0 | -1.0 | 0.5571 | 0.1414 | 0.7683 | 0.0284 | 0.0345 | 0.1151 | 0.8686 |
| 1.2882 | 2.6477 | 1300 | 0.5378 | 0.0764 | 0.0945 | 0.089 | -1.0 | -1.0 | 0.0852 | 0.3972 | 0.4941 | 0.5378 | -1.0 | -1.0 | 0.5378 | 0.1073 | 0.6603 | 0.034 | 0.0483 | 0.0878 | 0.9049 |
| 0.8294 | 2.7495 | 1350 | 0.5371 | 0.0767 | 0.0961 | 0.0866 | -1.0 | -1.0 | 0.0844 | 0.4005 | 0.5034 | 0.5468 | -1.0 | -1.0 | 0.5468 | 0.1278 | 0.6302 | 0.0068 | 0.0966 | 0.0955 | 0.9137 |
| 0.9082 | 2.8513 | 1400 | 0.5802 | 0.0719 | 0.0917 | 0.0829 | -1.0 | -1.0 | 0.0781 | 0.3829 | 0.4934 | 0.536 | -1.0 | -1.0 | 0.536 | 0.0747 | 0.654 | 0.0037 | 0.1138 | 0.1374 | 0.8402 |
| 0.8392 | 2.9532 | 1450 | 0.5597 | 0.0599 | 0.0715 | 0.0663 | -1.0 | -1.0 | 0.064 | 0.3819 | 0.5131 | 0.5684 | -1.0 | -1.0 | 0.5684 | 0.0673 | 0.7476 | 0.0009 | 0.131 | 0.1115 | 0.8265 |
| 0.601 | 3.0550 | 1500 | 0.5938 | 0.0492 | 0.0625 | 0.0532 | -1.0 | -1.0 | 0.0545 | 0.3377 | 0.4861 | 0.5589 | -1.0 | -1.0 | 0.5589 | 0.0563 | 0.7 | 0.0037 | 0.1207 | 0.0874 | 0.8559 |
| 0.7053 | 3.1568 | 1550 | 0.5777 | 0.0739 | 0.094 | 0.0844 | -1.0 | -1.0 | 0.0793 | 0.3408 | 0.4853 | 0.5219 | -1.0 | -1.0 | 0.5219 | 0.0683 | 0.5952 | 0.0343 | 0.0931 | 0.1192 | 0.8775 |
| 0.7158 | 3.2587 | 1600 | 0.5854 | 0.0637 | 0.0843 | 0.0666 | -1.0 | -1.0 | 0.0667 | 0.3329 | 0.4633 | 0.5151 | -1.0 | -1.0 | 0.5151 | 0.0646 | 0.5683 | 0.0048 | 0.0966 | 0.1218 | 0.8804 |
| 0.9677 | 3.3605 | 1650 | 0.5985 | 0.0989 | 0.1221 | 0.1134 | -1.0 | -1.0 | 0.1111 | 0.2947 | 0.4177 | 0.4762 | -1.0 | -1.0 | 0.4762 | 0.0786 | 0.4968 | 0.0018 | 0.0966 | 0.2164 | 0.8353 |
| 0.8341 | 3.4623 | 1700 | 0.5789 | 0.1197 | 0.1507 | 0.134 | -1.0 | -1.0 | 0.126 | 0.3459 | 0.4435 | 0.4582 | -1.0 | -1.0 | 0.4582 | 0.1337 | 0.4524 | 0.0003 | 0.0586 | 0.2251 | 0.8637 |
| 0.962 | 3.5642 | 1750 | 0.5207 | 0.11 | 0.1313 | 0.1236 | -1.0 | -1.0 | 0.1173 | 0.3525 | 0.4689 | 0.4927 | -1.0 | -1.0 | 0.4927 | 0.1182 | 0.5032 | 0.001 | 0.0966 | 0.2107 | 0.8784 |
| 0.4853 | 3.6660 | 1800 | 0.5804 | 0.1082 | 0.1377 | 0.1261 | -1.0 | -1.0 | 0.1116 | 0.3381 | 0.4678 | 0.489 | -1.0 | -1.0 | 0.489 | 0.1781 | 0.5048 | 0.0023 | 0.1241 | 0.1441 | 0.8382 |
| 0.9292 | 3.7678 | 1850 | 0.5584 | 0.0925 | 0.1167 | 0.1028 | -1.0 | -1.0 | 0.099 | 0.3454 | 0.4664 | 0.5041 | -1.0 | -1.0 | 0.5041 | 0.1293 | 0.5619 | 0.0063 | 0.0828 | 0.142 | 0.8676 |
| 0.552 | 3.8697 | 1900 | 0.5131 | 0.1002 | 0.1219 | 0.1123 | -1.0 | -1.0 | 0.1138 | 0.3719 | 0.4999 | 0.5265 | -1.0 | -1.0 | 0.5265 | 0.13 | 0.5841 | 0.0026 | 0.1276 | 0.168 | 0.8676 |
| 0.6468 | 3.9715 | 1950 | 0.5834 | 0.1186 | 0.1491 | 0.1313 | -1.0 | -1.0 | 0.1292 | 0.3901 | 0.49 | 0.5076 | -1.0 | -1.0 | 0.5076 | 0.1583 | 0.519 | 0.01 | 0.1655 | 0.1875 | 0.8382 |
| 0.925 | 4.0733 | 2000 | 0.4856 | 0.1184 | 0.1539 | 0.1284 | -1.0 | -1.0 | 0.1381 | 0.4197 | 0.5425 | 0.5677 | -1.0 | -1.0 | 0.5677 | 0.1504 | 0.6048 | 0.0149 | 0.1897 | 0.1899 | 0.9088 |
| 0.5978 | 4.1752 | 2050 | 0.4895 | 0.115 | 0.1406 | 0.1295 | -1.0 | -1.0 | 0.1282 | 0.3839 | 0.5032 | 0.5218 | -1.0 | -1.0 | 0.5218 | 0.1538 | 0.5587 | 0.0038 | 0.131 | 0.1875 | 0.8755 |
| 0.4953 | 4.2770 | 2100 | 0.4927 | 0.1009 | 0.1264 | 0.1104 | -1.0 | -1.0 | 0.1084 | 0.367 | 0.5038 | 0.5164 | -1.0 | -1.0 | 0.5164 | 0.1419 | 0.5 | 0.0081 | 0.1483 | 0.1528 | 0.901 |
| 0.8966 | 4.3788 | 2150 | 0.4611 | 0.1002 | 0.1254 | 0.1132 | -1.0 | -1.0 | 0.1154 | 0.3965 | 0.5181 | 0.5348 | -1.0 | -1.0 | 0.5348 | 0.1167 | 0.5429 | 0.0086 | 0.1655 | 0.1753 | 0.8961 |
| 0.5029 | 4.4807 | 2200 | 0.4622 | 0.1134 | 0.1431 | 0.1231 | -1.0 | -1.0 | 0.1239 | 0.4481 | 0.5523 | 0.5951 | -1.0 | -1.0 | 0.5951 | 0.1493 | 0.6857 | 0.0132 | 0.2103 | 0.1777 | 0.8892 |
| 0.791 | 4.5825 | 2250 | 0.5093 | 0.0874 | 0.1051 | 0.0997 | -1.0 | -1.0 | 0.0966 | 0.3781 | 0.4987 | 0.5458 | -1.0 | -1.0 | 0.5458 | 0.1002 | 0.6333 | 0.0007 | 0.1138 | 0.1613 | 0.8902 |
| 0.6545 | 4.6843 | 2300 | 0.4820 | 0.0976 | 0.1167 | 0.1102 | -1.0 | -1.0 | 0.1174 | 0.4407 | 0.5537 | 0.5862 | -1.0 | -1.0 | 0.5862 | 0.1573 | 0.6984 | 0.0012 | 0.1483 | 0.1344 | 0.9118 |
| 0.5782 | 4.7862 | 2350 | 0.4653 | 0.1146 | 0.1374 | 0.1307 | -1.0 | -1.0 | 0.1423 | 0.4482 | 0.5451 | 0.5855 | -1.0 | -1.0 | 0.5855 | 0.1619 | 0.6667 | 0.0034 | 0.2034 | 0.1785 | 0.8863 |
| 0.5309 | 4.8880 | 2400 | 0.4830 | 0.1009 | 0.1269 | 0.1202 | -1.0 | -1.0 | 0.1223 | 0.4266 | 0.5057 | 0.5419 | -1.0 | -1.0 | 0.5419 | 0.1643 | 0.6317 | 0.0015 | 0.1448 | 0.137 | 0.849 |
| 0.4851 | 4.9898 | 2450 | 0.4866 | 0.1464 | 0.1851 | 0.1672 | -1.0 | -1.0 | 0.1619 | 0.4202 | 0.5202 | 0.5719 | -1.0 | -1.0 | 0.5719 | 0.2452 | 0.6905 | 0.0029 | 0.1724 | 0.1909 | 0.8529 |
| 0.7152 | 5.0916 | 2500 | 0.4744 | 0.1459 | 0.1799 | 0.1648 | -1.0 | -1.0 | 0.1547 | 0.4379 | 0.5467 | 0.5785 | -1.0 | -1.0 | 0.5785 | 0.2351 | 0.7333 | 0.0018 | 0.169 | 0.2008 | 0.8333 |
| 0.2953 | 5.1935 | 2550 | 0.5260 | 0.1592 | 0.1982 | 0.1827 | -1.0 | -1.0 | 0.1695 | 0.4385 | 0.5087 | 0.5527 | -1.0 | -1.0 | 0.5527 | 0.2435 | 0.6365 | 0.0037 | 0.1793 | 0.2304 | 0.8422 |
| 0.5106 | 5.2953 | 2600 | 0.5279 | 0.1627 | 0.2051 | 0.18 | -1.0 | -1.0 | 0.1706 | 0.4105 | 0.4839 | 0.5344 | -1.0 | -1.0 | 0.5344 | 0.2688 | 0.619 | 0.001 | 0.1931 | 0.2183 | 0.7912 |
| 1.5618 | 5.3971 | 2650 | 0.4933 | 0.1624 | 0.1978 | 0.1818 | -1.0 | -1.0 | 0.1707 | 0.3791 | 0.4933 | 0.5232 | -1.0 | -1.0 | 0.5232 | 0.2336 | 0.573 | 0.0021 | 0.2241 | 0.2515 | 0.7725 |
| 0.6616 | 5.4990 | 2700 | 0.4607 | 0.1411 | 0.1741 | 0.1561 | -1.0 | -1.0 | 0.1554 | 0.3631 | 0.5007 | 0.5528 | -1.0 | -1.0 | 0.5528 | 0.1914 | 0.5476 | 0.0036 | 0.2793 | 0.2283 | 0.8314 |
| 1.5876 | 5.6008 | 2750 | 0.4688 | 0.1714 | 0.2055 | 0.1822 | -1.0 | -1.0 | 0.1853 | 0.426 | 0.5362 | 0.5845 | -1.0 | -1.0 | 0.5845 | 0.2096 | 0.6063 | 0.0069 | 0.3069 | 0.2976 | 0.8402 |
| 0.9064 | 5.7026 | 2800 | 0.4834 | 0.1512 | 0.1818 | 0.1695 | -1.0 | -1.0 | 0.1715 | 0.4667 | 0.5708 | 0.6274 | -1.0 | -1.0 | 0.6274 | 0.2307 | 0.6968 | 0.0076 | 0.3414 | 0.2153 | 0.8441 |
| 0.6631 | 5.8045 | 2850 | 0.4966 | 0.1668 | 0.2034 | 0.185 | -1.0 | -1.0 | 0.1879 | 0.4371 | 0.5178 | 0.5715 | -1.0 | -1.0 | 0.5715 | 0.2648 | 0.6349 | 0.0099 | 0.2483 | 0.2257 | 0.8314 |
| 2.0204 | 5.9063 | 2900 | 0.5232 | 0.1848 | 0.2317 | 0.1976 | -1.0 | -1.0 | 0.2007 | 0.4219 | 0.5219 | 0.5843 | -1.0 | -1.0 | 0.5843 | 0.3123 | 0.681 | 0.0237 | 0.3483 | 0.2183 | 0.7235 |
| 0.6208 | 6.0081 | 2950 | 0.4736 | 0.1749 | 0.2204 | 0.1895 | -1.0 | -1.0 | 0.1871 | 0.4238 | 0.5162 | 0.5837 | -1.0 | -1.0 | 0.5837 | 0.2772 | 0.646 | 0.016 | 0.3138 | 0.2315 | 0.7912 |
| 0.7104 | 6.1100 | 3000 | 0.4668 | 0.1787 | 0.2152 | 0.1945 | -1.0 | -1.0 | 0.1884 | 0.4139 | 0.5343 | 0.6019 | -1.0 | -1.0 | 0.6019 | 0.2652 | 0.6127 | 0.0075 | 0.3862 | 0.2635 | 0.8069 |
| 0.7477 | 6.2118 | 3050 | 0.5456 | 0.1533 | 0.1916 | 0.1728 | -1.0 | -1.0 | 0.1588 | 0.4022 | 0.5193 | 0.6083 | -1.0 | -1.0 | 0.6083 | 0.2289 | 0.5714 | 0.0248 | 0.4897 | 0.2062 | 0.7637 |
| 0.6839 | 6.3136 | 3100 | 0.5166 | 0.1163 | 0.1449 | 0.1254 | -1.0 | -1.0 | 0.1238 | 0.3623 | 0.5177 | 0.6251 | -1.0 | -1.0 | 0.6251 | 0.182 | 0.6349 | 0.0081 | 0.4552 | 0.1588 | 0.7853 |
| 1.0881 | 6.4155 | 3150 | 0.4868 | 0.1468 | 0.1806 | 0.1631 | -1.0 | -1.0 | 0.1538 | 0.3639 | 0.4961 | 0.5872 | -1.0 | -1.0 | 0.5872 | 0.201 | 0.5063 | 0.0072 | 0.4828 | 0.2324 | 0.7725 |
| 0.6136 | 6.5173 | 3200 | 0.5087 | 0.1154 | 0.1432 | 0.1284 | -1.0 | -1.0 | 0.1218 | 0.3638 | 0.5158 | 0.5888 | -1.0 | -1.0 | 0.5888 | 0.189 | 0.5794 | 0.0028 | 0.3724 | 0.1544 | 0.8147 |
| 0.7524 | 6.6191 | 3250 | 0.4734 | 0.153 | 0.1884 | 0.1643 | -1.0 | -1.0 | 0.1582 | 0.3827 | 0.509 | 0.6 | -1.0 | -1.0 | 0.6 | 0.2354 | 0.573 | 0.002 | 0.4034 | 0.2215 | 0.8235 |
| 0.6668 | 6.7210 | 3300 | 0.4848 | 0.1622 | 0.1973 | 0.1718 | -1.0 | -1.0 | 0.1668 | 0.3723 | 0.4999 | 0.591 | -1.0 | -1.0 | 0.591 | 0.2335 | 0.5127 | 0.0024 | 0.4172 | 0.2508 | 0.8431 |
| 0.6529 | 6.8228 | 3350 | 0.4945 | 0.117 | 0.1462 | 0.1282 | -1.0 | -1.0 | 0.1258 | 0.3225 | 0.4903 | 0.5723 | -1.0 | -1.0 | 0.5723 | 0.1648 | 0.5048 | 0.0032 | 0.3966 | 0.1829 | 0.8157 |
| 0.4778 | 6.9246 | 3400 | 0.5310 | 0.1211 | 0.1577 | 0.1307 | -1.0 | -1.0 | 0.1278 | 0.3232 | 0.4381 | 0.5086 | -1.0 | -1.0 | 0.5086 | 0.1466 | 0.481 | 0.0033 | 0.3379 | 0.2133 | 0.7069 |
| 0.6646 | 7.0265 | 3450 | 0.5276 | 0.125 | 0.1627 | 0.1357 | -1.0 | -1.0 | 0.1333 | 0.3163 | 0.43 | 0.5238 | -1.0 | -1.0 | 0.5238 | 0.1462 | 0.4508 | 0.0072 | 0.4 | 0.2216 | 0.7206 |
| 0.8104 | 7.1283 | 3500 | 0.5576 | 0.1225 | 0.1635 | 0.1285 | -1.0 | -1.0 | 0.132 | 0.3087 | 0.4179 | 0.511 | -1.0 | -1.0 | 0.511 | 0.1528 | 0.4603 | 0.007 | 0.4 | 0.2077 | 0.6725 |
| 0.801 | 7.2301 | 3550 | 0.5051 | 0.1465 | 0.1832 | 0.1588 | -1.0 | -1.0 | 0.1558 | 0.3237 | 0.4378 | 0.5166 | -1.0 | -1.0 | 0.5166 | 0.1797 | 0.4444 | 0.0098 | 0.3828 | 0.25 | 0.7225 |
| 0.5484 | 7.3320 | 3600 | 0.4944 | 0.1617 | 0.1932 | 0.1775 | -1.0 | -1.0 | 0.1667 | 0.3525 | 0.4632 | 0.5622 | -1.0 | -1.0 | 0.5622 | 0.2367 | 0.4841 | 0.0066 | 0.4241 | 0.2417 | 0.7784 |
| 0.5271 | 7.4338 | 3650 | 0.4661 | 0.1811 | 0.2237 | 0.1965 | -1.0 | -1.0 | 0.1862 | 0.3673 | 0.4609 | 0.5317 | -1.0 | -1.0 | 0.5317 | 0.2413 | 0.4635 | 0.011 | 0.3345 | 0.291 | 0.7971 |
| 0.4769 | 7.5356 | 3700 | 0.4799 | 0.206 | 0.2478 | 0.2217 | -1.0 | -1.0 | 0.2101 | 0.3542 | 0.4457 | 0.5043 | -1.0 | -1.0 | 0.5043 | 0.2343 | 0.427 | 0.014 | 0.3103 | 0.3699 | 0.7755 |
| 0.5822 | 7.6375 | 3750 | 0.4613 | 0.1984 | 0.2378 | 0.2212 | -1.0 | -1.0 | 0.2083 | 0.3518 | 0.4378 | 0.5114 | -1.0 | -1.0 | 0.5114 | 0.227 | 0.4413 | 0.0079 | 0.3448 | 0.3605 | 0.748 |
| 0.5706 | 7.7393 | 3800 | 0.4651 | 0.2286 | 0.2703 | 0.2447 | -1.0 | -1.0 | 0.2381 | 0.3655 | 0.452 | 0.5405 | -1.0 | -1.0 | 0.5405 | 0.2425 | 0.4714 | 0.013 | 0.4069 | 0.4301 | 0.7431 |
| 0.5366 | 7.8411 | 3850 | 0.4488 | 0.2597 | 0.3022 | 0.2783 | -1.0 | -1.0 | 0.266 | 0.4008 | 0.4565 | 0.5151 | -1.0 | -1.0 | 0.5151 | 0.3008 | 0.5254 | 0.0121 | 0.2897 | 0.4663 | 0.7304 |
| 0.3441 | 7.9430 | 3900 | 0.4445 | 0.2552 | 0.2979 | 0.2728 | -1.0 | -1.0 | 0.2594 | 0.4021 | 0.4654 | 0.5401 | -1.0 | -1.0 | 0.5401 | 0.3088 | 0.5524 | 0.0186 | 0.3345 | 0.4383 | 0.7333 |
| 0.3097 | 8.0448 | 3950 | 0.4632 | 0.2153 | 0.2686 | 0.2372 | -1.0 | -1.0 | 0.2228 | 0.3759 | 0.4661 | 0.5719 | -1.0 | -1.0 | 0.5719 | 0.2775 | 0.5635 | 0.0102 | 0.4207 | 0.3581 | 0.7314 |
| 0.5967 | 8.1466 | 4000 | 0.4446 | 0.2026 | 0.249 | 0.2207 | -1.0 | -1.0 | 0.2129 | 0.3703 | 0.4724 | 0.5804 | -1.0 | -1.0 | 0.5804 | 0.2722 | 0.5397 | 0.0129 | 0.4379 | 0.3227 | 0.7637 |
| 0.5614 | 8.2485 | 4050 | 0.4194 | 0.2358 | 0.2855 | 0.2567 | -1.0 | -1.0 | 0.2442 | 0.3982 | 0.4728 | 0.5751 | -1.0 | -1.0 | 0.5751 | 0.2696 | 0.4952 | 0.0156 | 0.4448 | 0.4222 | 0.7853 |
| 0.4889 | 8.3503 | 4100 | 0.4300 | 0.217 | 0.2592 | 0.2456 | -1.0 | -1.0 | 0.2235 | 0.394 | 0.4763 | 0.5915 | -1.0 | -1.0 | 0.5915 | 0.2868 | 0.5317 | 0.0051 | 0.4379 | 0.3592 | 0.8049 |
| 1.2673 | 8.4521 | 4150 | 0.4346 | 0.2151 | 0.2509 | 0.2369 | -1.0 | -1.0 | 0.2218 | 0.3905 | 0.482 | 0.5879 | -1.0 | -1.0 | 0.5879 | 0.2719 | 0.5222 | 0.0087 | 0.4552 | 0.3646 | 0.7863 |
| 0.6729 | 8.5540 | 4200 | 0.4171 | 0.2327 | 0.2748 | 0.255 | -1.0 | -1.0 | 0.2413 | 0.4106 | 0.4912 | 0.5866 | -1.0 | -1.0 | 0.5866 | 0.2814 | 0.5175 | 0.008 | 0.4207 | 0.4085 | 0.8216 |
| 0.7014 | 8.6558 | 4250 | 0.4043 | 0.2677 | 0.3162 | 0.3003 | -1.0 | -1.0 | 0.2748 | 0.4198 | 0.4855 | 0.5993 | -1.0 | -1.0 | 0.5993 | 0.3233 | 0.5222 | 0.0083 | 0.4552 | 0.4714 | 0.8206 |
| 0.7492 | 8.7576 | 4300 | 0.4004 | 0.2642 | 0.3078 | 0.2941 | -1.0 | -1.0 | 0.271 | 0.4279 | 0.5052 | 0.6121 | -1.0 | -1.0 | 0.6121 | 0.3225 | 0.5635 | 0.0066 | 0.4759 | 0.4634 | 0.7971 |
| 0.4239 | 8.8595 | 4350 | 0.4437 | 0.2589 | 0.3104 | 0.2932 | -1.0 | -1.0 | 0.266 | 0.3972 | 0.4808 | 0.5762 | -1.0 | -1.0 | 0.5762 | 0.2964 | 0.4873 | 0.0077 | 0.4483 | 0.4727 | 0.7931 |
| 0.5212 | 8.9613 | 4400 | 0.4309 | 0.2673 | 0.3196 | 0.2887 | -1.0 | -1.0 | 0.2705 | 0.4035 | 0.4836 | 0.5894 | -1.0 | -1.0 | 0.5894 | 0.3463 | 0.5333 | 0.0077 | 0.4966 | 0.4479 | 0.7382 |
| 0.8995 | 9.0631 | 4450 | 0.4414 | 0.2576 | 0.3093 | 0.2763 | -1.0 | -1.0 | 0.2595 | 0.4094 | 0.4847 | 0.5996 | -1.0 | -1.0 | 0.5996 | 0.3314 | 0.5508 | 0.017 | 0.5 | 0.4245 | 0.748 |
| 1.5476 | 9.1650 | 4500 | 0.4524 | 0.242 | 0.2912 | 0.2629 | -1.0 | -1.0 | 0.2489 | 0.4091 | 0.4879 | 0.5937 | -1.0 | -1.0 | 0.5937 | 0.3223 | 0.5556 | 0.0154 | 0.4931 | 0.3882 | 0.7324 |
| 0.7314 | 9.2668 | 4550 | 0.4297 | 0.2381 | 0.2907 | 0.2602 | -1.0 | -1.0 | 0.244 | 0.4097 | 0.501 | 0.5964 | -1.0 | -1.0 | 0.5964 | 0.3204 | 0.554 | 0.0224 | 0.5138 | 0.3715 | 0.7216 |
| 0.3865 | 9.3686 | 4600 | 0.4467 | 0.2645 | 0.314 | 0.2845 | -1.0 | -1.0 | 0.2701 | 0.4264 | 0.5071 | 0.6163 | -1.0 | -1.0 | 0.6163 | 0.3504 | 0.5905 | 0.02 | 0.4966 | 0.423 | 0.7618 |
| 0.3017 | 9.4705 | 4650 | 0.4504 | 0.282 | 0.3464 | 0.3033 | -1.0 | -1.0 | 0.2892 | 0.4396 | 0.5249 | 0.6318 | -1.0 | -1.0 | 0.6318 | 0.416 | 0.6921 | 0.0176 | 0.4897 | 0.4124 | 0.7137 |
| 0.7409 | 9.5723 | 4700 | 0.4369 | 0.301 | 0.356 | 0.3257 | -1.0 | -1.0 | 0.3071 | 0.4518 | 0.5085 | 0.6085 | -1.0 | -1.0 | 0.6085 | 0.4307 | 0.6571 | 0.0215 | 0.4724 | 0.4508 | 0.6961 |
| 1.5215 | 9.6741 | 4750 | 0.4673 | 0.3043 | 0.3455 | 0.3268 | -1.0 | -1.0 | 0.3056 | 0.4372 | 0.5032 | 0.6009 | -1.0 | -1.0 | 0.6009 | 0.4026 | 0.6095 | 0.0084 | 0.4862 | 0.5019 | 0.7069 |
| 0.3358 | 9.7760 | 4800 | 0.4813 | 0.3042 | 0.3514 | 0.3316 | -1.0 | -1.0 | 0.306 | 0.4493 | 0.4972 | 0.6007 | -1.0 | -1.0 | 0.6007 | 0.3894 | 0.5889 | 0.014 | 0.4828 | 0.5092 | 0.7304 |
| 0.9629 | 9.8778 | 4850 | 0.4473 | 0.3066 | 0.3599 | 0.3268 | -1.0 | -1.0 | 0.3105 | 0.4398 | 0.497 | 0.5878 | -1.0 | -1.0 | 0.5878 | 0.399 | 0.5984 | 0.0233 | 0.4552 | 0.4975 | 0.7098 |
| 0.524 | 9.9796 | 4900 | 0.4228 | 0.3062 | 0.3519 | 0.3275 | -1.0 | -1.0 | 0.3086 | 0.4226 | 0.4831 | 0.5544 | -1.0 | -1.0 | 0.5544 | 0.3748 | 0.554 | 0.0186 | 0.3897 | 0.5251 | 0.7196 |
| 0.6995 | 10.0815 | 4950 | 0.4831 | 0.2857 | 0.3441 | 0.318 | -1.0 | -1.0 | 0.2873 | 0.4042 | 0.4572 | 0.5353 | -1.0 | -1.0 | 0.5353 | 0.318 | 0.4825 | 0.0201 | 0.3724 | 0.5189 | 0.751 |
| 0.3898 | 10.1833 | 5000 | 0.4427 | 0.2371 | 0.2802 | 0.2588 | -1.0 | -1.0 | 0.2426 | 0.3619 | 0.4637 | 0.5534 | -1.0 | -1.0 | 0.5534 | 0.2811 | 0.4683 | 0.0103 | 0.4517 | 0.4199 | 0.7402 |
| 1.4225 | 10.2851 | 5050 | 0.4479 | 0.2618 | 0.3105 | 0.2844 | -1.0 | -1.0 | 0.2657 | 0.3895 | 0.4483 | 0.5483 | -1.0 | -1.0 | 0.5483 | 0.2747 | 0.4476 | 0.0189 | 0.4345 | 0.4917 | 0.7627 |
| 0.538 | 10.3870 | 5100 | 0.4383 | 0.2825 | 0.3342 | 0.3014 | -1.0 | -1.0 | 0.2861 | 0.3928 | 0.4584 | 0.5561 | -1.0 | -1.0 | 0.5561 | 0.288 | 0.4317 | 0.0161 | 0.4414 | 0.5432 | 0.7951 |
| 0.3099 | 10.4888 | 5150 | 0.4078 | 0.2763 | 0.3263 | 0.2933 | -1.0 | -1.0 | 0.2824 | 0.402 | 0.477 | 0.5678 | -1.0 | -1.0 | 0.5678 | 0.275 | 0.4556 | 0.0213 | 0.4517 | 0.5324 | 0.7961 |
| 0.4364 | 10.5906 | 5200 | 0.4022 | 0.287 | 0.3355 | 0.3045 | -1.0 | -1.0 | 0.2901 | 0.4022 | 0.4638 | 0.5534 | -1.0 | -1.0 | 0.5534 | 0.2937 | 0.473 | 0.0187 | 0.4 | 0.5487 | 0.7873 |
| 0.5209 | 10.6925 | 5250 | 0.4358 | 0.2923 | 0.3345 | 0.3092 | -1.0 | -1.0 | 0.2941 | 0.4067 | 0.463 | 0.5251 | -1.0 | -1.0 | 0.5251 | 0.2984 | 0.4571 | 0.0152 | 0.3759 | 0.5634 | 0.7422 |
| 0.3867 | 10.7943 | 5300 | 0.4289 | 0.2775 | 0.3189 | 0.2994 | -1.0 | -1.0 | 0.2802 | 0.3961 | 0.4691 | 0.5462 | -1.0 | -1.0 | 0.5462 | 0.2774 | 0.454 | 0.0131 | 0.4345 | 0.5419 | 0.75 |
| 0.6977 | 10.8961 | 5350 | 0.3995 | 0.2918 | 0.3262 | 0.3128 | -1.0 | -1.0 | 0.2926 | 0.4155 | 0.4672 | 0.5683 | -1.0 | -1.0 | 0.5683 | 0.3237 | 0.5032 | 0.012 | 0.4724 | 0.5398 | 0.7294 |
| 0.3935 | 10.9980 | 5400 | 0.4075 | 0.304 | 0.3499 | 0.3286 | -1.0 | -1.0 | 0.3054 | 0.4403 | 0.4865 | 0.6049 | -1.0 | -1.0 | 0.6049 | 0.3784 | 0.5921 | 0.017 | 0.4931 | 0.5166 | 0.7294 |
| 0.3714 | 11.0998 | 5450 | 0.4077 | 0.2963 | 0.3365 | 0.3237 | -1.0 | -1.0 | 0.2983 | 0.4231 | 0.4831 | 0.6061 | -1.0 | -1.0 | 0.6061 | 0.3805 | 0.5873 | 0.0091 | 0.531 | 0.4992 | 0.7 |
| 1.6946 | 11.2016 | 5500 | 0.4363 | 0.3013 | 0.3373 | 0.3221 | -1.0 | -1.0 | 0.3021 | 0.4214 | 0.4836 | 0.5916 | -1.0 | -1.0 | 0.5916 | 0.3619 | 0.5571 | 0.0071 | 0.5069 | 0.5349 | 0.7108 |
| 0.329 | 11.3035 | 5550 | 0.4442 | 0.2945 | 0.3349 | 0.3152 | -1.0 | -1.0 | 0.2964 | 0.3984 | 0.4753 | 0.5868 | -1.0 | -1.0 | 0.5868 | 0.3299 | 0.5333 | 0.0164 | 0.5103 | 0.5373 | 0.7167 |
| 0.6864 | 11.4053 | 5600 | 0.4718 | 0.2981 | 0.3363 | 0.3213 | -1.0 | -1.0 | 0.3016 | 0.4057 | 0.466 | 0.5568 | -1.0 | -1.0 | 0.5568 | 0.3361 | 0.5159 | 0.0082 | 0.4241 | 0.5501 | 0.7304 |
| 0.5806 | 11.5071 | 5650 | 0.4585 | 0.2923 | 0.333 | 0.313 | -1.0 | -1.0 | 0.2993 | 0.3925 | 0.4592 | 0.5627 | -1.0 | -1.0 | 0.5627 | 0.3377 | 0.519 | 0.0069 | 0.4621 | 0.5323 | 0.7069 |
| 0.4736 | 11.6090 | 5700 | 0.4330 | 0.2974 | 0.3369 | 0.3147 | -1.0 | -1.0 | 0.302 | 0.3923 | 0.4525 | 0.5617 | -1.0 | -1.0 | 0.5617 | 0.3266 | 0.4873 | 0.0165 | 0.4655 | 0.549 | 0.7324 |
| 0.633 | 11.7108 | 5750 | 0.4385 | 0.2962 | 0.3312 | 0.3188 | -1.0 | -1.0 | 0.3 | 0.3825 | 0.4397 | 0.5478 | -1.0 | -1.0 | 0.5478 | 0.3222 | 0.4794 | 0.0076 | 0.4552 | 0.5588 | 0.7088 |
| 1.298 | 11.8126 | 5800 | 0.4444 | 0.3013 | 0.3393 | 0.323 | -1.0 | -1.0 | 0.3038 | 0.4001 | 0.4537 | 0.5502 | -1.0 | -1.0 | 0.5502 | 0.3416 | 0.5063 | 0.0096 | 0.4414 | 0.5528 | 0.7029 |
| 0.8351 | 11.9145 | 5850 | 0.4634 | 0.3017 | 0.3379 | 0.324 | -1.0 | -1.0 | 0.3082 | 0.3907 | 0.4491 | 0.5365 | -1.0 | -1.0 | 0.5365 | 0.3151 | 0.4746 | 0.0088 | 0.4034 | 0.5812 | 0.7314 |
| 1.0177 | 12.0163 | 5900 | 0.4514 | 0.3039 | 0.334 | 0.3228 | -1.0 | -1.0 | 0.3096 | 0.4035 | 0.4511 | 0.5477 | -1.0 | -1.0 | 0.5477 | 0.3202 | 0.4841 | 0.0064 | 0.4207 | 0.5853 | 0.7382 |
| 1.3682 | 12.1181 | 5950 | 0.4470 | 0.3057 | 0.3383 | 0.3269 | -1.0 | -1.0 | 0.3108 | 0.4154 | 0.4693 | 0.5624 | -1.0 | -1.0 | 0.5624 | 0.311 | 0.454 | 0.0069 | 0.4724 | 0.5993 | 0.7608 |
| 0.8238 | 12.2200 | 6000 | 0.4072 | 0.2973 | 0.327 | 0.3156 | -1.0 | -1.0 | 0.3013 | 0.3962 | 0.4594 | 0.5594 | -1.0 | -1.0 | 0.5594 | 0.3261 | 0.4698 | 0.0069 | 0.4828 | 0.5588 | 0.7255 |
| 0.6402 | 12.3218 | 6050 | 0.4230 | 0.3007 | 0.3494 | 0.3217 | -1.0 | -1.0 | 0.3043 | 0.3922 | 0.4391 | 0.5483 | -1.0 | -1.0 | 0.5483 | 0.3108 | 0.4333 | 0.0178 | 0.4793 | 0.5735 | 0.7324 |
| 0.537 | 12.4236 | 6100 | 0.4037 | 0.3244 | 0.3624 | 0.347 | -1.0 | -1.0 | 0.3287 | 0.4309 | 0.4829 | 0.5771 | -1.0 | -1.0 | 0.5771 | 0.3955 | 0.5698 | 0.0125 | 0.4448 | 0.5652 | 0.7167 |
| 0.4765 | 12.5255 | 6150 | 0.4080 | 0.33 | 0.3676 | 0.3505 | -1.0 | -1.0 | 0.3343 | 0.4206 | 0.4785 | 0.5532 | -1.0 | -1.0 | 0.5532 | 0.3935 | 0.5667 | 0.0143 | 0.3655 | 0.5821 | 0.7275 |
| 1.7486 | 12.6273 | 6200 | 0.4253 | 0.3357 | 0.3728 | 0.3523 | -1.0 | -1.0 | 0.3413 | 0.4408 | 0.4794 | 0.5461 | -1.0 | -1.0 | 0.5461 | 0.4151 | 0.5984 | 0.0193 | 0.3103 | 0.5726 | 0.7294 |
| 0.3369 | 12.7291 | 6250 | 0.4018 | 0.3303 | 0.3768 | 0.3485 | -1.0 | -1.0 | 0.3348 | 0.4193 | 0.4688 | 0.5343 | -1.0 | -1.0 | 0.5343 | 0.3943 | 0.5667 | 0.0199 | 0.3207 | 0.5766 | 0.7157 |
| 0.4465 | 12.8310 | 6300 | 0.4211 | 0.3166 | 0.3477 | 0.3357 | -1.0 | -1.0 | 0.3214 | 0.4098 | 0.4517 | 0.5413 | -1.0 | -1.0 | 0.5413 | 0.3691 | 0.5286 | 0.0099 | 0.3759 | 0.5709 | 0.7196 |
| 0.3394 | 12.9328 | 6350 | 0.4136 | 0.3198 | 0.3612 | 0.3356 | -1.0 | -1.0 | 0.3227 | 0.408 | 0.4577 | 0.5278 | -1.0 | -1.0 | 0.5278 | 0.3652 | 0.5111 | 0.0218 | 0.3448 | 0.5725 | 0.7275 |
| 0.4738 | 13.0346 | 6400 | 0.4102 | 0.3177 | 0.3612 | 0.3376 | -1.0 | -1.0 | 0.3228 | 0.4211 | 0.4689 | 0.5562 | -1.0 | -1.0 | 0.5562 | 0.3793 | 0.5492 | 0.0165 | 0.3862 | 0.5573 | 0.7333 |
| 1.1177 | 13.1365 | 6450 | 0.3955 | 0.311 | 0.3477 | 0.3333 | -1.0 | -1.0 | 0.3149 | 0.414 | 0.4836 | 0.5618 | -1.0 | -1.0 | 0.5618 | 0.3778 | 0.5667 | 0.0124 | 0.4 | 0.5427 | 0.7186 |
| 0.4671 | 13.2383 | 6500 | 0.3912 | 0.3208 | 0.3562 | 0.3411 | -1.0 | -1.0 | 0.3235 | 0.4182 | 0.4794 | 0.5495 | -1.0 | -1.0 | 0.5495 | 0.3871 | 0.5492 | 0.017 | 0.3759 | 0.5584 | 0.7235 |
| 0.4744 | 13.3401 | 6550 | 0.3923 | 0.3147 | 0.3527 | 0.3375 | -1.0 | -1.0 | 0.3184 | 0.4197 | 0.4639 | 0.5525 | -1.0 | -1.0 | 0.5525 | 0.3776 | 0.5476 | 0.0109 | 0.3862 | 0.5555 | 0.7235 |
| 0.5302 | 13.4420 | 6600 | 0.3963 | 0.3306 | 0.3744 | 0.3502 | -1.0 | -1.0 | 0.3364 | 0.4402 | 0.4883 | 0.5607 | -1.0 | -1.0 | 0.5607 | 0.4113 | 0.6063 | 0.0178 | 0.3483 | 0.5627 | 0.7275 |
| 0.3128 | 13.5438 | 6650 | 0.3925 | 0.3071 | 0.3531 | 0.3291 | -1.0 | -1.0 | 0.3108 | 0.4132 | 0.4631 | 0.5608 | -1.0 | -1.0 | 0.5608 | 0.3564 | 0.519 | 0.026 | 0.4517 | 0.5388 | 0.7118 |
| 0.3991 | 13.6456 | 6700 | 0.3757 | 0.3115 | 0.3548 | 0.3292 | -1.0 | -1.0 | 0.315 | 0.4058 | 0.4678 | 0.5529 | -1.0 | -1.0 | 0.5529 | 0.3718 | 0.5222 | 0.0191 | 0.4207 | 0.5435 | 0.7157 |
| 0.4445 | 13.7475 | 6750 | 0.3741 | 0.3134 | 0.3565 | 0.3357 | -1.0 | -1.0 | 0.3174 | 0.4255 | 0.4833 | 0.5707 | -1.0 | -1.0 | 0.5707 | 0.3946 | 0.581 | 0.018 | 0.431 | 0.5274 | 0.7 |
| 0.3567 | 13.8493 | 6800 | 0.3990 | 0.3027 | 0.3473 | 0.3238 | -1.0 | -1.0 | 0.3079 | 0.4061 | 0.465 | 0.5409 | -1.0 | -1.0 | 0.5409 | 0.3903 | 0.5841 | 0.0174 | 0.369 | 0.5003 | 0.6696 |
| 0.4034 | 13.9511 | 6850 | 0.3865 | 0.3056 | 0.348 | 0.3279 | -1.0 | -1.0 | 0.3117 | 0.4194 | 0.4654 | 0.5516 | -1.0 | -1.0 | 0.5516 | 0.3961 | 0.5873 | 0.0159 | 0.3862 | 0.5049 | 0.6814 |
| 0.5333 | 14.0530 | 6900 | 0.3983 | 0.3128 | 0.3539 | 0.3331 | -1.0 | -1.0 | 0.3159 | 0.4222 | 0.4709 | 0.5709 | -1.0 | -1.0 | 0.5709 | 0.4218 | 0.6238 | 0.0165 | 0.431 | 0.5001 | 0.6578 |
| 0.7354 | 14.1548 | 6950 | 0.4141 | 0.3272 | 0.372 | 0.3492 | -1.0 | -1.0 | 0.3285 | 0.4241 | 0.4748 | 0.5782 | -1.0 | -1.0 | 0.5782 | 0.4271 | 0.6079 | 0.0103 | 0.4552 | 0.5441 | 0.6716 |
| 1.6592 | 14.2566 | 7000 | 0.3964 | 0.3168 | 0.3669 | 0.332 | -1.0 | -1.0 | 0.3187 | 0.4248 | 0.4698 | 0.5514 | -1.0 | -1.0 | 0.5514 | 0.4006 | 0.5825 | 0.0296 | 0.4138 | 0.5203 | 0.6578 |
| 0.6544 | 14.3585 | 7050 | 0.4193 | 0.2988 | 0.3352 | 0.3178 | -1.0 | -1.0 | 0.3063 | 0.4099 | 0.4889 | 0.5591 | -1.0 | -1.0 | 0.5591 | 0.3766 | 0.5952 | 0.015 | 0.4172 | 0.5047 | 0.6647 |
| 0.5879 | 14.4603 | 7100 | 0.4189 | 0.2819 | 0.3277 | 0.2934 | -1.0 | -1.0 | 0.2917 | 0.3982 | 0.4751 | 0.5659 | -1.0 | -1.0 | 0.5659 | 0.3546 | 0.5778 | 0.0216 | 0.4897 | 0.4694 | 0.6304 |
| 0.9165 | 14.5621 | 7150 | 0.3850 | 0.298 | 0.3386 | 0.3164 | -1.0 | -1.0 | 0.3051 | 0.4053 | 0.4772 | 0.5795 | -1.0 | -1.0 | 0.5795 | 0.3774 | 0.5952 | 0.0191 | 0.5 | 0.4977 | 0.6431 |
| 0.4671 | 14.6640 | 7200 | 0.4131 | 0.2873 | 0.3272 | 0.3003 | -1.0 | -1.0 | 0.2908 | 0.3774 | 0.4611 | 0.5565 | -1.0 | -1.0 | 0.5565 | 0.371 | 0.5825 | 0.0109 | 0.4517 | 0.4799 | 0.6353 |
| 0.3135 | 14.7658 | 7250 | 0.3958 | 0.3117 | 0.357 | 0.3311 | -1.0 | -1.0 | 0.3136 | 0.4233 | 0.4766 | 0.5582 | -1.0 | -1.0 | 0.5582 | 0.3805 | 0.5651 | 0.0177 | 0.4172 | 0.537 | 0.6922 |
| 0.2984 | 14.8676 | 7300 | 0.4165 | 0.2985 | 0.3395 | 0.3232 | -1.0 | -1.0 | 0.2995 | 0.4044 | 0.4586 | 0.5298 | -1.0 | -1.0 | 0.5298 | 0.3353 | 0.4905 | 0.0154 | 0.3931 | 0.5447 | 0.7059 |
| 0.5092 | 14.9695 | 7350 | 0.4012 | 0.3097 | 0.3525 | 0.3261 | -1.0 | -1.0 | 0.3108 | 0.4195 | 0.4751 | 0.5544 | -1.0 | -1.0 | 0.5544 | 0.3381 | 0.4952 | 0.0212 | 0.4414 | 0.5697 | 0.7265 |
| 0.5818 | 15.0713 | 7400 | 0.4022 | 0.323 | 0.3556 | 0.3451 | -1.0 | -1.0 | 0.325 | 0.4258 | 0.4885 | 0.569 | -1.0 | -1.0 | 0.569 | 0.3931 | 0.5794 | 0.0113 | 0.4138 | 0.5648 | 0.7137 |
| 0.6533 | 15.1731 | 7450 | 0.3960 | 0.3241 | 0.3741 | 0.3378 | -1.0 | -1.0 | 0.3257 | 0.4284 | 0.4922 | 0.5554 | -1.0 | -1.0 | 0.5554 | 0.3926 | 0.5746 | 0.0229 | 0.3828 | 0.5567 | 0.7088 |
| 0.6955 | 15.2749 | 7500 | 0.4087 | 0.3219 | 0.3635 | 0.3385 | -1.0 | -1.0 | 0.3243 | 0.4285 | 0.4747 | 0.554 | -1.0 | -1.0 | 0.554 | 0.3785 | 0.5619 | 0.0271 | 0.4 | 0.5602 | 0.7 |
| 0.8639 | 15.3768 | 7550 | 0.3846 | 0.3315 | 0.3748 | 0.3533 | -1.0 | -1.0 | 0.3325 | 0.4372 | 0.4959 | 0.5614 | -1.0 | -1.0 | 0.5614 | 0.4088 | 0.5841 | 0.0225 | 0.4 | 0.563 | 0.7 |
| 1.3757 | 15.4786 | 7600 | 0.3801 | 0.325 | 0.3725 | 0.3428 | -1.0 | -1.0 | 0.3261 | 0.423 | 0.4748 | 0.5633 | -1.0 | -1.0 | 0.5633 | 0.3912 | 0.5651 | 0.0261 | 0.4414 | 0.5575 | 0.6833 |
| 0.3398 | 15.5804 | 7650 | 0.3954 | 0.3275 | 0.3848 | 0.3504 | -1.0 | -1.0 | 0.3278 | 0.426 | 0.4721 | 0.556 | -1.0 | -1.0 | 0.556 | 0.3895 | 0.5587 | 0.0312 | 0.4103 | 0.5619 | 0.699 |
| 0.3513 | 15.6823 | 7700 | 0.3872 | 0.3471 | 0.4016 | 0.3683 | -1.0 | -1.0 | 0.3472 | 0.4347 | 0.4777 | 0.5524 | -1.0 | -1.0 | 0.5524 | 0.4352 | 0.5952 | 0.0259 | 0.369 | 0.5801 | 0.6931 |
| 0.3229 | 15.7841 | 7750 | 0.3868 | 0.345 | 0.3949 | 0.365 | -1.0 | -1.0 | 0.345 | 0.4324 | 0.4707 | 0.5523 | -1.0 | -1.0 | 0.5523 | 0.4224 | 0.5698 | 0.0252 | 0.3862 | 0.5875 | 0.701 |
| 0.4801 | 15.8859 | 7800 | 0.3777 | 0.3469 | 0.3939 | 0.3674 | -1.0 | -1.0 | 0.3469 | 0.4322 | 0.4742 | 0.5558 | -1.0 | -1.0 | 0.5558 | 0.4443 | 0.5968 | 0.0232 | 0.3862 | 0.5731 | 0.6843 |
| 0.435 | 15.9878 | 7850 | 0.3780 | 0.341 | 0.396 | 0.3624 | -1.0 | -1.0 | 0.341 | 0.4307 | 0.4605 | 0.5433 | -1.0 | -1.0 | 0.5433 | 0.4309 | 0.581 | 0.0234 | 0.3655 | 0.5687 | 0.6833 |
| 0.7239 | 16.0896 | 7900 | 0.3660 | 0.3447 | 0.3955 | 0.3645 | -1.0 | -1.0 | 0.3456 | 0.425 | 0.4812 | 0.5628 | -1.0 | -1.0 | 0.5628 | 0.4352 | 0.581 | 0.0207 | 0.4172 | 0.5782 | 0.6902 |
| 1.2456 | 16.1914 | 7950 | 0.3725 | 0.3472 | 0.3957 | 0.3688 | -1.0 | -1.0 | 0.3479 | 0.4266 | 0.4852 | 0.5588 | -1.0 | -1.0 | 0.5588 | 0.4215 | 0.5524 | 0.0158 | 0.3966 | 0.6044 | 0.7275 |
| 0.2957 | 16.2933 | 8000 | 0.3768 | 0.3507 | 0.4022 | 0.3706 | -1.0 | -1.0 | 0.3508 | 0.4378 | 0.4779 | 0.5526 | -1.0 | -1.0 | 0.5526 | 0.4157 | 0.5397 | 0.0236 | 0.3759 | 0.6127 | 0.7422 |
| 0.4006 | 16.3951 | 8050 | 0.3907 | 0.3475 | 0.3892 | 0.3687 | -1.0 | -1.0 | 0.3475 | 0.4274 | 0.476 | 0.5599 | -1.0 | -1.0 | 0.5599 | 0.4305 | 0.5651 | 0.0158 | 0.3931 | 0.5961 | 0.7216 |
| 0.4983 | 16.4969 | 8100 | 0.3968 | 0.3424 | 0.3891 | 0.3656 | -1.0 | -1.0 | 0.3425 | 0.4373 | 0.479 | 0.5618 | -1.0 | -1.0 | 0.5618 | 0.4233 | 0.5746 | 0.019 | 0.4069 | 0.5849 | 0.7039 |
| 0.2923 | 16.5988 | 8150 | 0.3845 | 0.3447 | 0.3961 | 0.3655 | -1.0 | -1.0 | 0.3447 | 0.4211 | 0.4838 | 0.5712 | -1.0 | -1.0 | 0.5712 | 0.4258 | 0.5762 | 0.0234 | 0.4345 | 0.585 | 0.7029 |
| 0.84 | 16.7006 | 8200 | 0.3824 | 0.3363 | 0.3864 | 0.3554 | -1.0 | -1.0 | 0.3363 | 0.4302 | 0.4862 | 0.569 | -1.0 | -1.0 | 0.569 | 0.4107 | 0.5635 | 0.0244 | 0.4483 | 0.5738 | 0.6951 |
| 0.2858 | 16.8024 | 8250 | 0.3763 | 0.333 | 0.3824 | 0.3473 | -1.0 | -1.0 | 0.3332 | 0.4283 | 0.4733 | 0.5618 | -1.0 | -1.0 | 0.5618 | 0.4072 | 0.554 | 0.0287 | 0.4414 | 0.5629 | 0.6902 |
| 0.4264 | 16.9043 | 8300 | 0.3812 | 0.3304 | 0.3742 | 0.3496 | -1.0 | -1.0 | 0.3311 | 0.4134 | 0.4662 | 0.5397 | -1.0 | -1.0 | 0.5397 | 0.4157 | 0.5556 | 0.0241 | 0.3931 | 0.5515 | 0.6706 |
| 0.3899 | 17.0061 | 8350 | 0.3761 | 0.3397 | 0.3917 | 0.354 | -1.0 | -1.0 | 0.3406 | 0.4323 | 0.4779 | 0.5549 | -1.0 | -1.0 | 0.5549 | 0.4234 | 0.5714 | 0.0333 | 0.4138 | 0.5625 | 0.6794 |
| 0.5517 | 17.1079 | 8400 | 0.3705 | 0.3284 | 0.3807 | 0.343 | -1.0 | -1.0 | 0.329 | 0.4243 | 0.4733 | 0.5515 | -1.0 | -1.0 | 0.5515 | 0.4095 | 0.5524 | 0.0308 | 0.4138 | 0.5448 | 0.6882 |
| 0.544 | 17.2098 | 8450 | 0.3792 | 0.336 | 0.3857 | 0.3524 | -1.0 | -1.0 | 0.3368 | 0.4312 | 0.4852 | 0.5726 | -1.0 | -1.0 | 0.5726 | 0.4437 | 0.6127 | 0.0233 | 0.4414 | 0.541 | 0.6637 |
| 0.3121 | 17.3116 | 8500 | 0.3753 | 0.332 | 0.3765 | 0.3514 | -1.0 | -1.0 | 0.3329 | 0.4281 | 0.4854 | 0.5728 | -1.0 | -1.0 | 0.5728 | 0.4091 | 0.5524 | 0.0207 | 0.4621 | 0.5662 | 0.7039 |
| 0.3784 | 17.4134 | 8550 | 0.3753 | 0.3253 | 0.3734 | 0.346 | -1.0 | -1.0 | 0.3262 | 0.4168 | 0.4705 | 0.5613 | -1.0 | -1.0 | 0.5613 | 0.4143 | 0.5524 | 0.0272 | 0.4552 | 0.5346 | 0.6765 |
| 0.5904 | 17.5153 | 8600 | 0.3805 | 0.3321 | 0.3801 | 0.3479 | -1.0 | -1.0 | 0.333 | 0.4223 | 0.4804 | 0.57 | -1.0 | -1.0 | 0.57 | 0.4165 | 0.5524 | 0.0239 | 0.4655 | 0.5559 | 0.6922 |
| 0.9018 | 17.6171 | 8650 | 0.3653 | 0.3388 | 0.383 | 0.356 | -1.0 | -1.0 | 0.3396 | 0.4253 | 0.4848 | 0.563 | -1.0 | -1.0 | 0.563 | 0.4224 | 0.554 | 0.0234 | 0.431 | 0.5708 | 0.7039 |
| 0.5026 | 17.7189 | 8700 | 0.3699 | 0.3383 | 0.39 | 0.3527 | -1.0 | -1.0 | 0.3388 | 0.4107 | 0.4615 | 0.5512 | -1.0 | -1.0 | 0.5512 | 0.418 | 0.5397 | 0.0297 | 0.4276 | 0.5671 | 0.6863 |
| 0.4129 | 17.8208 | 8750 | 0.3559 | 0.3404 | 0.3899 | 0.3565 | -1.0 | -1.0 | 0.3407 | 0.417 | 0.4661 | 0.5523 | -1.0 | -1.0 | 0.5523 | 0.4173 | 0.5397 | 0.0274 | 0.4241 | 0.5766 | 0.6931 |
| 1.2525 | 17.9226 | 8800 | 0.3814 | 0.344 | 0.3841 | 0.364 | -1.0 | -1.0 | 0.344 | 0.4183 | 0.4756 | 0.5572 | -1.0 | -1.0 | 0.5572 | 0.4258 | 0.554 | 0.018 | 0.4069 | 0.5882 | 0.7108 |
| 0.8418 | 18.0244 | 8850 | 0.3730 | 0.3514 | 0.4029 | 0.3701 | -1.0 | -1.0 | 0.3514 | 0.4266 | 0.4815 | 0.5654 | -1.0 | -1.0 | 0.5654 | 0.4392 | 0.5651 | 0.0306 | 0.4241 | 0.5845 | 0.7069 |
| 0.495 | 18.1263 | 8900 | 0.3763 | 0.3444 | 0.3905 | 0.3612 | -1.0 | -1.0 | 0.3444 | 0.4186 | 0.4725 | 0.5656 | -1.0 | -1.0 | 0.5656 | 0.4318 | 0.554 | 0.0277 | 0.4517 | 0.5736 | 0.6912 |
| 0.36 | 18.2281 | 8950 | 0.3530 | 0.3521 | 0.4019 | 0.368 | -1.0 | -1.0 | 0.3521 | 0.4288 | 0.482 | 0.5705 | -1.0 | -1.0 | 0.5705 | 0.4369 | 0.5683 | 0.0334 | 0.4414 | 0.5861 | 0.702 |
| 0.4354 | 18.3299 | 9000 | 0.3601 | 0.3503 | 0.4012 | 0.3664 | -1.0 | -1.0 | 0.3503 | 0.4312 | 0.4799 | 0.5787 | -1.0 | -1.0 | 0.5787 | 0.4436 | 0.581 | 0.0306 | 0.4621 | 0.5767 | 0.6931 |
| 0.5433 | 18.4318 | 9050 | 0.3590 | 0.3488 | 0.3922 | 0.3697 | -1.0 | -1.0 | 0.349 | 0.4301 | 0.4779 | 0.6067 | -1.0 | -1.0 | 0.6067 | 0.4526 | 0.5968 | 0.0203 | 0.5379 | 0.5736 | 0.6853 |
| 0.6011 | 18.5336 | 9100 | 0.3560 | 0.3511 | 0.4051 | 0.3698 | -1.0 | -1.0 | 0.3515 | 0.4314 | 0.4811 | 0.6133 | -1.0 | -1.0 | 0.6133 | 0.4526 | 0.5968 | 0.0289 | 0.5586 | 0.5717 | 0.6843 |
| 0.2583 | 18.6354 | 9150 | 0.3576 | 0.3511 | 0.4031 | 0.3684 | -1.0 | -1.0 | 0.3514 | 0.4279 | 0.4781 | 0.6068 | -1.0 | -1.0 | 0.6068 | 0.4429 | 0.581 | 0.031 | 0.5483 | 0.5794 | 0.6912 |
| 0.5591 | 18.7373 | 9200 | 0.3600 | 0.3502 | 0.4005 | 0.3676 | -1.0 | -1.0 | 0.3503 | 0.4289 | 0.481 | 0.6006 | -1.0 | -1.0 | 0.6006 | 0.4438 | 0.581 | 0.0291 | 0.5345 | 0.5778 | 0.6863 |
| 0.4953 | 18.8391 | 9250 | 0.3594 | 0.3566 | 0.4109 | 0.3753 | -1.0 | -1.0 | 0.3566 | 0.4336 | 0.4857 | 0.6063 | -1.0 | -1.0 | 0.6063 | 0.459 | 0.5968 | 0.0304 | 0.531 | 0.5804 | 0.6912 |
| 0.4214 | 18.9409 | 9300 | 0.3589 | 0.3629 | 0.4123 | 0.3825 | -1.0 | -1.0 | 0.363 | 0.4404 | 0.4896 | 0.616 | -1.0 | -1.0 | 0.616 | 0.4711 | 0.6111 | 0.0314 | 0.5379 | 0.5863 | 0.699 |
| 0.3492 | 19.0428 | 9350 | 0.3619 | 0.3687 | 0.4228 | 0.385 | -1.0 | -1.0 | 0.3687 | 0.4475 | 0.5016 | 0.6258 | -1.0 | -1.0 | 0.6258 | 0.4855 | 0.6286 | 0.0326 | 0.5448 | 0.588 | 0.7039 |
| 0.2945 | 19.1446 | 9400 | 0.3604 | 0.3687 | 0.4216 | 0.3893 | -1.0 | -1.0 | 0.3687 | 0.4473 | 0.5009 | 0.6273 | -1.0 | -1.0 | 0.6273 | 0.4842 | 0.6238 | 0.0331 | 0.5552 | 0.5886 | 0.7029 |
| 0.4994 | 19.2464 | 9450 | 0.3693 | 0.3645 | 0.4175 | 0.3822 | -1.0 | -1.0 | 0.3645 | 0.4383 | 0.4914 | 0.596 | -1.0 | -1.0 | 0.596 | 0.4677 | 0.5968 | 0.0347 | 0.4862 | 0.5911 | 0.7049 |
| 0.6958 | 19.3483 | 9500 | 0.3723 | 0.367 | 0.4224 | 0.3906 | -1.0 | -1.0 | 0.367 | 0.4442 | 0.4995 | 0.6076 | -1.0 | -1.0 | 0.6076 | 0.475 | 0.6095 | 0.0348 | 0.5034 | 0.5913 | 0.7098 |
| 1.15 | 19.4501 | 9550 | 0.3687 | 0.3686 | 0.424 | 0.393 | -1.0 | -1.0 | 0.3686 | 0.446 | 0.5074 | 0.5925 | -1.0 | -1.0 | 0.5925 | 0.4795 | 0.6095 | 0.0256 | 0.4483 | 0.6008 | 0.7196 |
| 0.2574 | 19.5519 | 9600 | 0.3728 | 0.3677 | 0.4216 | 0.3886 | -1.0 | -1.0 | 0.3677 | 0.4432 | 0.5001 | 0.5851 | -1.0 | -1.0 | 0.5851 | 0.4751 | 0.5984 | 0.0244 | 0.4345 | 0.6036 | 0.7225 |
| 0.6085 | 19.6538 | 9650 | 0.3694 | 0.37 | 0.4247 | 0.3921 | -1.0 | -1.0 | 0.37 | 0.4478 | 0.5026 | 0.5923 | -1.0 | -1.0 | 0.5923 | 0.4797 | 0.6095 | 0.0263 | 0.4448 | 0.6041 | 0.7225 |
| 0.5051 | 19.7556 | 9700 | 0.3746 | 0.3675 | 0.4224 | 0.3897 | -1.0 | -1.0 | 0.3676 | 0.4438 | 0.4982 | 0.5856 | -1.0 | -1.0 | 0.5856 | 0.4792 | 0.6095 | 0.0257 | 0.4345 | 0.5976 | 0.7127 |
| 0.9448 | 19.8574 | 9750 | 0.3722 | 0.3666 | 0.4219 | 0.3883 | -1.0 | -1.0 | 0.3666 | 0.4438 | 0.5005 | 0.5948 | -1.0 | -1.0 | 0.5948 | 0.478 | 0.6095 | 0.0249 | 0.4621 | 0.5968 | 0.7127 |
| 0.4886 | 19.9593 | 9800 | 0.3704 | 0.3673 | 0.4248 | 0.3885 | -1.0 | -1.0 | 0.3673 | 0.4389 | 0.5011 | 0.6011 | -1.0 | -1.0 | 0.6011 | 0.4772 | 0.6111 | 0.0256 | 0.4793 | 0.599 | 0.7127 |
| 0.1613 | 20.0611 | 9850 | 0.3702 | 0.3704 | 0.4271 | 0.3914 | -1.0 | -1.0 | 0.3704 | 0.4433 | 0.5049 | 0.5934 | -1.0 | -1.0 | 0.5934 | 0.478 | 0.6095 | 0.0264 | 0.4483 | 0.6067 | 0.7225 |
| 0.3438 | 20.1629 | 9900 | 0.3700 | 0.3701 | 0.428 | 0.3914 | -1.0 | -1.0 | 0.3701 | 0.4419 | 0.5035 | 0.5931 | -1.0 | -1.0 | 0.5931 | 0.4787 | 0.6095 | 0.0258 | 0.4483 | 0.6057 | 0.7216 |
| 0.6573 | 20.2648 | 9950 | 0.3702 | 0.37 | 0.4245 | 0.3912 | -1.0 | -1.0 | 0.37 | 0.4491 | 0.5049 | 0.5934 | -1.0 | -1.0 | 0.5934 | 0.4783 | 0.6095 | 0.0253 | 0.4483 | 0.6063 | 0.7225 |
| 0.5515 | 20.3666 | 10000 | 0.3703 | 0.3703 | 0.4284 | 0.3912 | -1.0 | -1.0 | 0.3703 | 0.4491 | 0.5038 | 0.5934 | -1.0 | -1.0 | 0.5934 | 0.4782 | 0.6095 | 0.0261 | 0.4483 | 0.6064 | 0.7225 |
### Framework versions
- Transformers 4.51.3
- Pytorch 2.7.0+cu126
- Datasets 3.5.0
- Tokenizers 0.21.1
|
dzanbek/c2145cfe-eadf-4b88-bbb3-9d1792fc61c2
|
dzanbek
| 2025-04-28T05:23:55Z | 0 | 0 |
peft
|
[
"peft",
"safetensors",
"llama",
"axolotl",
"generated_from_trainer",
"base_model:01-ai/Yi-1.5-9B-Chat-16K",
"base_model:adapter:01-ai/Yi-1.5-9B-Chat-16K",
"license:apache-2.0",
"8-bit",
"bitsandbytes",
"region:us"
] | null | 2025-04-28T05:05:35Z |
---
library_name: peft
license: apache-2.0
base_model: 01-ai/Yi-1.5-9B-Chat-16K
tags:
- axolotl
- generated_from_trainer
model-index:
- name: c2145cfe-eadf-4b88-bbb3-9d1792fc61c2
results: []
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
[<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl)
<details><summary>See axolotl config</summary>
axolotl version: `0.4.1`
```yaml
absolute_data_files: false
adapter: lora
base_model: 01-ai/Yi-1.5-9B-Chat-16K
bf16: true
chat_template: llama3
dataset_prepared_path: /workspace/axolotl
datasets:
- data_files:
- 5a632c5faf4d9e56_train_data.json
ds_type: json
format: custom
path: /workspace/input_data/5a632c5faf4d9e56_train_data.json
type:
field_input: document_title
field_instruction: question
field_output: answer
format: '{instruction} {input}'
no_input_format: '{instruction}'
system_format: '{system}'
system_prompt: ''
debug: null
deepspeed: null
early_stopping_patience: null
eval_max_new_tokens: 128
eval_table_size: null
evals_per_epoch: 1
flash_attention: true
fp16: null
fsdp: null
fsdp_config: null
gradient_accumulation_steps: 1
gradient_checkpointing: true
gradient_clipping: 0.5
group_by_length: false
hub_model_id: dzanbek/c2145cfe-eadf-4b88-bbb3-9d1792fc61c2
hub_repo: null
hub_strategy: end
hub_token: null
learning_rate: 5.0e-06
load_in_4bit: false
load_in_8bit: true
local_rank: null
logging_steps: 1
lora_alpha: 64
lora_dropout: 0.05
lora_fan_in_fan_out: null
lora_model_dir: null
lora_r: 32
lora_target_linear: true
lr_scheduler: cosine
max_steps: 200
micro_batch_size: 8
mixed_precision: bf16
mlflow_experiment_name: /tmp/5a632c5faf4d9e56_train_data.json
model_type: AutoModelForCausalLM
num_epochs: 1
optimizer: adamw_bnb_8bit
output_dir: miner_id_24
pad_to_sequence_len: true
resume_from_checkpoint: null
s2_attention: null
sample_packing: false
saves_per_epoch: 1
sequence_len: 1024
strict: false
tf32: false
tokenizer_type: AutoTokenizer
train_on_inputs: false
trust_remote_code: true
val_set_size: 0.05
wandb_entity: null
wandb_mode: online
wandb_name: 5e837649-8f38-4a30-ade2-a231d08208ee
wandb_project: s56-2
wandb_run: your_name
wandb_runid: 5e837649-8f38-4a30-ade2-a231d08208ee
warmup_steps: 5
weight_decay: 0.01
xformers_attention: true
```
</details><br>
# c2145cfe-eadf-4b88-bbb3-9d1792fc61c2
This model is a fine-tuned version of [01-ai/Yi-1.5-9B-Chat-16K](https://huggingface.co/01-ai/Yi-1.5-9B-Chat-16K) on the None dataset.
It achieves the following results on the evaluation set:
- Loss: 2.0123
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 5e-06
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- optimizer: Use OptimizerNames.ADAMW_BNB with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: cosine
- lr_scheduler_warmup_steps: 5
- training_steps: 200
### Training results
| Training Loss | Epoch | Step | Validation Loss |
|:-------------:|:------:|:----:|:---------------:|
| 2.4051 | 0.0596 | 200 | 2.0123 |
### Framework versions
- PEFT 0.13.2
- Transformers 4.46.0
- Pytorch 2.5.0+cu124
- Datasets 3.0.1
- Tokenizers 0.20.1
|
sunkichen/embodied_reasoner_qwen2vl
|
sunkichen
| 2025-04-28T05:23:54Z | 0 | 0 |
transformers
|
[
"transformers",
"safetensors",
"qwen2_vl",
"image-text-to-text",
"llama-factory",
"full",
"generated_from_trainer",
"conversational",
"base_model:Qwen/Qwen2-VL-7B-Instruct",
"base_model:finetune:Qwen/Qwen2-VL-7B-Instruct",
"license:other",
"text-generation-inference",
"endpoints_compatible",
"region:us"
] |
image-text-to-text
| 2025-04-28T05:19:00Z |
---
library_name: transformers
license: other
base_model: Qwen/Qwen2-VL-7B-Instruct
tags:
- llama-factory
- full
- generated_from_trainer
model-index:
- name: qwen2_vl_ir65536_ds_embodied_o1_ct16384_lr1d0e-5_pbs1_g8_e1d0
results: []
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# qwen2_vl_ir65536_ds_embodied_o1_ct16384_lr1d0e-5_pbs1_g8_e1d0
This model is a fine-tuned version of [Qwen/Qwen2-VL-7B-Instruct](https://huggingface.co/Qwen/Qwen2-VL-7B-Instruct) on the embodied_o1 dataset.
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 1e-05
- train_batch_size: 1
- eval_batch_size: 8
- seed: 42
- distributed_type: multi-GPU
- num_devices: 2
- gradient_accumulation_steps: 8
- total_train_batch_size: 16
- total_eval_batch_size: 16
- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: cosine
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 1.0
### Training results
### Framework versions
- Transformers 4.49.0
- Pytorch 2.6.0+cu124
- Datasets 3.3.2
- Tokenizers 0.21.0
|
mlfoundations-dev/d1_science_gpt_0.3k
|
mlfoundations-dev
| 2025-04-28T05:23:25Z | 0 | 0 |
transformers
|
[
"transformers",
"tensorboard",
"safetensors",
"qwen2",
"text-generation",
"llama-factory",
"full",
"generated_from_trainer",
"conversational",
"base_model:Qwen/Qwen2.5-7B-Instruct",
"base_model:finetune:Qwen/Qwen2.5-7B-Instruct",
"license:other",
"autotrain_compatible",
"text-generation-inference",
"endpoints_compatible",
"region:us"
] |
text-generation
| 2025-04-28T05:20:51Z |
---
library_name: transformers
license: other
base_model: Qwen/Qwen2.5-7B-Instruct
tags:
- llama-factory
- full
- generated_from_trainer
model-index:
- name: d1_science_gpt_0.3k
results: []
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# d1_science_gpt_0.3k
This model is a fine-tuned version of [Qwen/Qwen2.5-7B-Instruct](https://huggingface.co/Qwen/Qwen2.5-7B-Instruct) on the mlfoundations-dev/d1_science_gpt_0.3k dataset.
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 1e-05
- train_batch_size: 1
- eval_batch_size: 8
- seed: 42
- distributed_type: multi-GPU
- num_devices: 16
- gradient_accumulation_steps: 2
- total_train_batch_size: 32
- total_eval_batch_size: 128
- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: cosine
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 13.0
### Training results
### Framework versions
- Transformers 4.46.1
- Pytorch 2.6.0a0+ecf3bae40a.nv25.01
- Datasets 3.5.0
- Tokenizers 0.20.3
|
robiulawaldev/6181a855-2b32-454f-b853-58d412b06348
|
robiulawaldev
| 2025-04-28T05:22:02Z | 0 | 0 |
transformers
|
[
"transformers",
"generated_from_trainer",
"endpoints_compatible",
"region:us"
] | null | 2025-04-28T05:21:36Z |
---
library_name: transformers
model_name: robiulawaldev/6181a855-2b32-454f-b853-58d412b06348
tags:
- generated_from_trainer
licence: license
---
# Model Card for robiulawaldev/6181a855-2b32-454f-b853-58d412b06348
This model is a fine-tuned version of [None](https://huggingface.co/None).
It has been trained using [TRL](https://github.com/huggingface/trl).
## Quick start
```python
from transformers import pipeline
question = "If you had a time machine, but could only go to the past or the future once and never return, which would you choose and why?"
generator = pipeline("text-generation", model="None", device="cuda")
output = generator([{"role": "user", "content": question}], max_new_tokens=128, return_full_text=False)[0]
print(output["generated_text"])
```
### Framework versions
- TRL: 0.12.0
- Transformers: 4.46.3
- Pytorch: 2.5.1
- Datasets: 3.1.0
- Tokenizers: 0.20.3
## Citations
Cite DPO as:
```bibtex
@inproceedings{rafailov2023direct,
title = {{Direct Preference Optimization: Your Language Model is Secretly a Reward Model}},
author = {Rafael Rafailov and Archit Sharma and Eric Mitchell and Christopher D. Manning and Stefano Ermon and Chelsea Finn},
year = 2023,
booktitle = {Advances in Neural Information Processing Systems 36: Annual Conference on Neural Information Processing Systems 2023, NeurIPS 2023, New Orleans, LA, USA, December 10 - 16, 2023},
url = {http://papers.nips.cc/paper_files/paper/2023/hash/a85b405ed65c6477a4fe8302b5e06ce7-Abstract-Conference.html},
editor = {Alice Oh and Tristan Naumann and Amir Globerson and Kate Saenko and Moritz Hardt and Sergey Levine},
}
```
Cite TRL as:
```bibtex
@misc{vonwerra2022trl,
title = {{TRL: Transformer Reinforcement Learning}},
author = {Leandro von Werra and Younes Belkada and Lewis Tunstall and Edward Beeching and Tristan Thrush and Nathan Lambert and Shengyi Huang and Kashif Rasul and Quentin Gallouédec},
year = 2020,
journal = {GitHub repository},
publisher = {GitHub},
howpublished = {\url{https://github.com/huggingface/trl}}
}
```
|
hcaeryks/bert-crf-harem
|
hcaeryks
| 2025-04-28T05:21:25Z | 126 | 0 | null |
[
"pytorch",
"safetensors",
"BERT_CRF",
"token-classification",
"custom_code",
"pt",
"dataset:harem",
"license:apache-2.0",
"region:us"
] |
token-classification
| 2025-04-15T00:53:49Z |
---
inference: False
license: apache-2.0
datasets:
- harem
language:
- pt
metrics:
- f1
pipeline_tag: token-classification
---
# Portuguese NER BERT-CRF HAREM Default
This model is a fine-tuned BERT model adapted for Named Entity Recognition (NER) tasks. It utilizes Conditional Random Fields (CRF) as the decoder.
The model follows the HAREM Default labeling scheme for NER. Additionally, it provides options for HAREM Selective and Conll-2003 labeling schemes.
## How to Use
You can employ this model using the Transformers library's *pipeline* for NER, or incorporate it as a conventional Transformer in the HuggingFace ecosystem.
```python
from transformers import pipeline
import torch
import nltk
ner_classifier = pipeline(
"ner",
model="arubenruben/NER-PT-BERT-CRF-HAREM-Default",
device=torch.device("cuda:0") if torch.cuda.is_available() else torch.device("cpu"),
trust_remote_code=True
)
text = "FCPorto vence o Benfica por 5-0 no Estádio do Dragão"
tokens = nltk.wordpunct_tokenize(text)
result = ner_classifier(tokens)
```
## Demo
There is a [Notebook](https://github.com/arubenruben/PT-Pump-Up/blob/master/BERT-CRF.ipynb) available to test our code.
## PT-Pump-Up
This model is integrated in the project [PT-Pump-Up](https://github.com/arubenruben/PT-Pump-Up)
## Evaluation
#### Testing Data
The model was tested on the Miniharem Testset.
### Results
F1-Score: 0.787
## Citation
Citation will be made available soon.
**BibTeX:**
:(
|
belyakoff/puzzle-search-model
|
belyakoff
| 2025-04-28T05:20:43Z | 0 | 0 |
sentence-transformers
|
[
"sentence-transformers",
"pytorch",
"safetensors",
"xlm-roberta",
"sentence-similarity",
"feature-extraction",
"generated_from_trainer",
"dataset_size:1413",
"loss:GISTEmbedLoss",
"arxiv:1908.10084",
"arxiv:2402.16829",
"base_model:intfloat/multilingual-e5-large-instruct",
"base_model:finetune:intfloat/multilingual-e5-large-instruct",
"autotrain_compatible",
"text-embeddings-inference",
"endpoints_compatible",
"region:us"
] |
sentence-similarity
| 2025-04-28T03:51:03Z |
---
tags:
- sentence-transformers
- sentence-similarity
- feature-extraction
- generated_from_trainer
- dataset_size:1413
- loss:GISTEmbedLoss
base_model: intfloat/multilingual-e5-large-instruct
widget:
- source_sentence: '1. проверить , запущен ли процесс fox. Если запущен, закрыть
2. открыть страницу в браузере. Адрес: avito.ru. Если будут напоминания для пользователя
— согласится на все.
3. навести мышь на меню Каталоги, дождаться появления подменю и навести мышь на
меню «каталог автомобилей». Кликнуть мышкой
4. найти все слова, которые выделены тэгом <h3>. Из списка сделать словарь, ключ
- текст, значение — другие параметры в тэге.
5. преобразовать словарь в датафрейм
6. сгруппировать данные датафрейма. Если есть повторы в ключах, сделать один ключ,
но с объединенным значением
7. выгрузить датафрем обратно в словарь.
8. Отправить сообщение в whatsapp об удачном завершении процесса'
sentences:
- 'Уведомление пользователя. describe: Выводит на экран всплывающее окно с указанным
текстом. Приостанавливает работу алгаритма до нажатия ''ok''. Окно закрывается
по нажатию кнопки ''ok''..Блок выводит на экран всплывающее окно с указанным текстом.
Окно закрывается по нажатию кнопки ‘Закрыть’. Выполнение алгоритма приостанавливается
во время отображения уведомления. Чтобы скопировать содержимое уведомления, необходимо
нажать кнопку “Скопировать”..Показать сообщений пользователю. Показывает в отдельном
окне.'
- "Добавить строку в DataFrame. describe: Добавляет строку в dataframe по номеру..Блок\
\ позволяет добавить, перезаписать и удалить строку в DataFrame. При этом необходимо\
\ указать сам DataFrame, строку для добавления/перезаписи и номер позиции..поскольку\
\ DataFrame (датафрейм) это таблица, к ней можно добавить строчку, удалить строчку\
\ или изменить строчку\nПри добавлении строки в датафрейм, нужно указать в какое\
\ место ее нужно поместить. \nПри удалении строки тоже нужно указать номер удаляемой\
\ строки.\nИ при изменении строки тоже указывается номер строки\nПри добавлении\
\ или изменении строки указывается список значений всех колонок\nПример\nДобавить\
\ строку в датафрейм dataframe\nУдалить строку из таблицы\nИзменить строку в датафрейме"
- 'Запущен ли процесс. describe: Проверяет, запущен ли указанный процесс. Если хотя
бы один экземпляр процесса запущен, то вернет True, иначе - False..Блок проверяет,
запущен ли указанный процесс. Если хотя бы один экземпляр процесса запущен, то
вернет “истина”, иначе вернет “ложь”.."Запущен ли процесс" — это вопрос, касающийся
текущего состояния процесса в операционной системе, обозначающий его активность
или присутствие в системе. Процесс считается запущенным, если он был инициирован
и выполняется в данный момент времени. Для проверки, запущен ли процесс, операционная
система использует таблицы процессов, где каждый процесс имеет свой уникальный
идентификатор (PID) и информацию о его статусе. Когда процесс находится в активном
состоянии, он использует ресурсы системы, такие как CPU и память, для выполнения
своих задач. Проверку статуса процесса можно осуществить с помощью различных инструментов,
например, с помощью командной строки или графических интерфейсов, отображающих
активные процессы.'
- source_sentence: "1. открыть Excel\n2. перейти на страницу «Итого»\n3. переместиться\
\ на кнопку «Стрелка вправо»\n4. прокрутить колесико мышки ровно 4 секунды, со\
\ скоростью 30 пикселей в секунду\n5. дальше выбор. Если в центре экрана видна\
\ печать NASA (пример есть в файле nasa.logo), то в переменную param записать\
\ Истина\n6. если в ровно в центре экрана нет печати NASA, то в переменную param\
\ записать None (не путать с ложь)\n7. если в переменной param записано None,\
\ найти в папке dir файл Roscosmos.data.\n8. открыть , прочитать все строки. Преобразовать\
\ строки в таблицу. Поставить фильтр на первой колонке , равенство, Роскосомос.\
\ \n9. полученную таблицу распечатать на принтере Printer1"
sentences:
- 'Удалить дубликаты. describe: Удаляет повторяющиеся элементы списка. Возвращает
список уникальных элементов..Блок удаляет повторяющиеся элементы списка, возвращает
список уникальных элементов..Аналог функции list(set()) в python. Удаляет дубли.
Сначала делаем множество, где одинаковые элементы объединяются, а потом из множества
делаем список
Примеры:
Удалить дубли
Получить уникальные значение'
- "Фильтровать табличные данные. describe: Фильтрует табличные данные по указанному\
\ столбцу и операции..Блок позволяет Фильтровать табличные данные. Необходимо\
\ указать:Таблицу - Путь к файлу или DataFrame источника данных;Столбец или список\
\ столбцов - Столбец или список столбцов для фильтрации;Операция - Операция сравнения\
\ для фильтрации;Значение - Значение или список значений для фильтрации;Движок\
\ обработки данных - Определяет способ использования ресурсов процессора при обработке\
\ таблиц.Для визуальной работы с данными, кликните по кнопке:Визуализация данныхФорма\
\ визуализации данных - встроенный инструмент Puzzle RPA, который позволяет загружать\
\ и просматривать различные наборы данных..датафрейм, как таблица, может быть\
\ использован для быстрой фильтрации. Можем наложить отбор на любую колонку и\
\ посмотреть что останется\nЗначение в колонке , на которую накладывается отбор\
\ может проверяться на:\nРавенству какому либо значению\nНе равенству\nНа меньше\n\
Больше\nЕсли в колонке есть пустые значение\nНаоборот, выбрать те, где заполнено\
\ \nВ списке значений\nПримеры\nОтфильтровать данные по колонке\nОставить только\
\ те строки, где..\nНаложить отбор на датафрейм"
- 'Изменить порядок. describe: Блок меняет порядок списка на обратный..Инвертирует
строку. Последний символ становится первым. А первый последним.
Например
Инвертировать строку «полисад». Ответ «дасилоп».'
- source_sentence: Скопировать число в файле Excel в столбце "Количество в граммах".
Извлечь данные из буфера обмена и выполнить деление этого числа на 1000. Вызвать
через командную строку калькулятор и на калькуляторе возвести в квадрат результат
деления
sentences:
- 'Прочитать письма. describe: Считывает письма электронной почты с указанными параметрами..Блок
позволяет прочитать письма по IMAP.Требуется указать:Данные почтового аккаунта,
который будет прочитан;Адрес сервера;Папку для сохранения вложений из писем.Дополнительно
нажатием на “+” можно добавлять следующие параметры:Дату, с которой получать сообщения;Дату,
до которой получать сообщения;Отправителя;Получателя;Тему сообщения;Тело сообщения;Подстроку
в теме или теле сообщения;ID-сообщения;Наличие вложения;Наличие флага;Получить
только не прочитанные;Отметить сообщение прочитанным;Отметить сообщение флагом.Некоторые
почтовые сервисы не поддерживают работу всех фильтров..процесс получения и отображения
содержания электронного письма, отправленного через почтовую службу, в почтовом
клиенте или веб-интерфейсе. Он включает в себя доступ к почтовому ящику, выбор
конкретного письма и его открытие для просмотра. При этом письма могут содержать
текст, вложенные файлы, изображения и ссылки, которые пользователь может просмотреть.
Во время чтения письма происходит декодирование и отображение данных, полученных
с почтового сервера. Этот процесс может быть выполнен как на компьютере, так и
на мобильных устройствах через специализированные приложения или веб-сайты'
- 'Остановить секундомер. describe: Останавливет секундомер и сохраняет результат
в переменню.Блок останавливает секундомер и сохраняет результат в переменную.
Единица измерения времени - секунды..действие, заключающееся в прекращении отсчёта
времени, фиксируемого устройством, предназначенным для измерения интервалов времени.
Обычно, секундомер активируется нажатием кнопки, и его остановка происходит также
нажатием на соответствующую кнопку или команду. После остановки секундомер фиксирует
текущий результат в виде времени, прошедшего с начала отсчёта. Остановка может
быть выполнена вручную или автоматически, в зависимости от типа устройства. После
остановки можно записать результат или повторить отсчёт времени, начиная новый
цикл'
- 'Арифметические операции. describe: В блоке есть два паза для добавления чисел,
кликнув по текущему условию, можно выбрать операцию, которую требуется произвести
с числами. Блок имеет выпадающее меню. Клик по символу раскрывающегося списка
открывает следующее меню:В меню представлены следующие опции:+ -возвращает сумму
двух чисел;-- возвращает разность двух чисел;×- возвращает произведение двух чисел;÷
-возвращает частное от деления первого числа на второе;^- возвращает первое число,
возведенное в степень второго..Нужен для
Сложения (+)
Вычитания (-)
Умножения (*)
Деления (/)
Возведение в степень (^)
двух чисел.
Например:
Сложить два числа
Найти остаток
5*9 = 45
1-8=-7
2:2 = 1
Увеличить число на 8'
- source_sentence: '1. подключиться к базе данных Postgres. Параметры подключение
взять из глобальных переменных
2. таблица Date, выбрать все даты прошлого года (list1)
3. таблица Numbers, выбрать все числа, которые не делятся на 2 (list2)
4. все даты в list1 преобразовать в строки в формате YyYy:Dd:Hhhh
5. для всех чисел list2 найти остаток от деления на 5.
6. объединить оба списка в один list3. Сохранить список в текстовый файл file.txt
7. проверить, если логин пароль для доступа на сайт my_fork.fr
8. если нет, то добавить с логином ME паролем 123dfg
9. загрузить file.txt на сайт my_fork.fr'
sentences:
- 'Сделать скриншот. describe: Сохраняет в файл скриншот всего экрана..Блок сохраняет
в файл скриншот всего экрана. Требуется указать путь к файлу с указанием названия
и расширения файла (.png). Файл будет создан автоматически по указанному пути..процесс
создания цифровой копии изображения или изображения и текста, отображаемых на
экране компьютера или другого устройства, такого как смартфон или планшет. Эта
операция позволяет сохранить текущее состояние дисплея в виде файла, который может
быть использован для различных целей, включая демонстрацию ошибок программного
обеспечения, сохранение важной информации или обмен изображениями через интернет.
Скриншоты обычно сохраняются в форматах изображений, таких как PNG, JPEG или BMP.
Для создания скриншота используются встроенные средства операционной системы,
специализированное программное обеспечение или горячие клавиши на клавиатуре.
Полученные скриншоты могут быть редактированы с помощью графических редакторов
для выделения важных элементов или добавления комментариев перед тем, как их использовать'
- 'Триггер по письму. describe: Ждет появления определенного сообщения в электронной
почте..Блок ожидает появление определенного письма в электронной почте.Требуется
указать:Данные почтового аккаунта, который будет прочитан;Адрес сервера;Время
ожидания.Дополнительно нажатием на “+” можно добавлять следующие параметры:Отправителя;Получателя;Тему
сообщения;Тело сообщения;Подстроку в теме или теле сообщения;Наличие вложения;Отметить
сообщение прочитанным;Отметить сообщение флагом.Некоторые почтовые сервисы не
поддерживают работу всех фильтров..автоматическое событие или условие, которое
активируется при получении нового письма на электронную почту. Этот триггер может
быть настроен для различных действий, например, отправки уведомлений, переноса
письма в определённую папку или запуска скрипта. Он работает на основе заданных
критериев, таких как отправитель, тема письма или ключевые слова в содержимом.
Триггер может быть реализован в почтовых клиентах или с помощью серверных автоматизаций,
например, через API почтовых сервисов. Основной целью является автоматизация обработки
входящих сообщений без необходимости вручную отслеживать каждое письмо'
- 'Остаток от деления. describe: Блок возвращает остаток от деления двух чисел..Математическая
операция , которая получает остаток от деления двух чисел
В python это операция %.
Например
Найти остаток от деления 15 на 3. Ответ 0
Найти остаток от деления 15 на 10. Ответ 5'
- source_sentence: с помощью bash скрипта узнать все рабочие процессы. В цикле начать
их обходить. Если процесс начинается на цифру, то остановить его. В файле delete_processes.txt
дописать имя закрытого процесса
sentences:
- 'Переключиться на процесс. describe: Блок позволяет подключиться к запущенному
процессу «1С», для дальнейшего взаимодействия с программой..Если толстый клиент
1с открыт, но был свернут, этот блок может вернуть в фокус 1с предприятие. '
- 'Дописать в файл. describe: Дописывает текст в конец указанного текстового или
json-файла..Блок дописывает текст в конец указанного текстового или json-файла..В
конец текстового файла с расширением txt или json дописать текст
Примеры
Добавить в текстовый файл
Дописать текст в файл'
- 'Прочитать из Word. describe: Считывает содержимое указанного документа Word.
Возвращает считанные данные в виде строки..Блок считывает содержимое указанного
файла Word, Поддерживаемый формат файла - docx. Возвращает строку, в строке содержатся
данные форматирования..ворд это текстовый документ, с возможностью форматирования
текста. Этот текст можно прочитать в переменную и потом обрабатывать текст. Укажите
путь к word файлу и файл будет прочитан
Примеры
Прочитать ворд
Получить текст из word файла'
pipeline_tag: sentence-similarity
library_name: sentence-transformers
---
# SentenceTransformer based on intfloat/multilingual-e5-large-instruct
This is a [sentence-transformers](https://www.SBERT.net) model finetuned from [intfloat/multilingual-e5-large-instruct](https://huggingface.co/intfloat/multilingual-e5-large-instruct). It maps sentences & paragraphs to a 1024-dimensional dense vector space and can be used for semantic textual similarity, semantic search, paraphrase mining, text classification, clustering, and more.
## Model Details
### Model Description
- **Model Type:** Sentence Transformer
- **Base model:** [intfloat/multilingual-e5-large-instruct](https://huggingface.co/intfloat/multilingual-e5-large-instruct) <!-- at revision 84344a23ee1820ac951bc365f1e91d094a911763 -->
- **Maximum Sequence Length:** 512 tokens
- **Output Dimensionality:** 1024 dimensions
- **Similarity Function:** Cosine Similarity
<!-- - **Training Dataset:** Unknown -->
<!-- - **Language:** Unknown -->
<!-- - **License:** Unknown -->
### Model Sources
- **Documentation:** [Sentence Transformers Documentation](https://sbert.net)
- **Repository:** [Sentence Transformers on GitHub](https://github.com/UKPLab/sentence-transformers)
- **Hugging Face:** [Sentence Transformers on Hugging Face](https://huggingface.co/models?library=sentence-transformers)
### Full Model Architecture
```
SentenceTransformer(
(0): Transformer({'max_seq_length': 512, 'do_lower_case': False}) with Transformer model: XLMRobertaModel
(1): Pooling({'word_embedding_dimension': 1024, 'pooling_mode_cls_token': False, 'pooling_mode_mean_tokens': True, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False, 'include_prompt': True})
(2): Normalize()
)
```
## Usage
### Direct Usage (Sentence Transformers)
First install the Sentence Transformers library:
```bash
pip install -U sentence-transformers
```
Then you can load this model and run inference.
```python
from sentence_transformers import SentenceTransformer
# Download from the 🤗 Hub
model = SentenceTransformer("belyakoff/puzzle-search-model")
# Run inference
sentences = [
'с помощью bash скрипта узнать все рабочие процессы. В цикле начать их обходить. Если процесс начинается на цифру, то остановить его. В файле delete_processes.txt дописать имя закрытого процесса',
'Дописать в файл. describe: Дописывает текст в конец указанного текстового или json-файла..Блок дописывает текст в конец указанного текстового или json-файла..В конец текстового файла с расширением txt или json дописать текст\nПримеры\nДобавить в текстовый файл\nДописать текст в файл',
'Переключиться на процесс. describe: Блок позволяет подключиться к запущенному процессу «1С», для дальнейшего взаимодействия с программой..Если толстый клиент 1с открыт, но был свернут, этот блок может вернуть в фокус 1с предприятие. ',
]
embeddings = model.encode(sentences)
print(embeddings.shape)
# [3, 1024]
# Get the similarity scores for the embeddings
similarities = model.similarity(embeddings, embeddings)
print(similarities.shape)
# [3, 3]
```
<!--
### Direct Usage (Transformers)
<details><summary>Click to see the direct usage in Transformers</summary>
</details>
-->
<!--
### Downstream Usage (Sentence Transformers)
You can finetune this model on your own dataset.
<details><summary>Click to expand</summary>
</details>
-->
<!--
### Out-of-Scope Use
*List how the model may foreseeably be misused and address what users ought not to do with the model.*
-->
<!--
## Bias, Risks and Limitations
*What are the known or foreseeable issues stemming from this model? You could also flag here known failure cases or weaknesses of the model.*
-->
<!--
### Recommendations
*What are recommendations with respect to the foreseeable issues? For example, filtering explicit content.*
-->
## Training Details
### Training Dataset
#### Unnamed Dataset
* Size: 1,413 training samples
* Columns: <code>anchor</code> and <code>positive</code>
* Approximate statistics based on the first 1000 samples:
| | anchor | positive |
|:--------|:------------------------------------------------------------------------------------|:-------------------------------------------------------------------------------------|
| type | string | string |
| details | <ul><li>min: 15 tokens</li><li>mean: 82.45 tokens</li><li>max: 326 tokens</li></ul> | <ul><li>min: 42 tokens</li><li>mean: 216.75 tokens</li><li>max: 512 tokens</li></ul> |
* Samples:
| anchor | positive |
|:----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|:-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
| <code>1. Авторизация в 1С-веб<br>2. Переключиться на страницу Файлы. Перейти в раздел документации.<br>3. Скачивание файла, сохранение на диск с проверкой SSL-сертификата, если это предусмотрено параметром `ssl_verify`.<br>4. Применение алгоритма сжатия к PDF-файлу, с конвертацией в оттенки серого и сохранением текстового слоя.<br>5. Сохранения сжатого файла в заданную директорию по пути `directory_path`.<br>6. уведомление об успешном скачивании, сжатии PDF-файла и сохранении с указанием размеров файлов до и после операции сжатия.</code> | <code>Добавить фильтр 1С-веб. describe: Позволяет выбрать один или несколько вариантов для открытия страницы/переключения на страницу..Блок “Добавить фильтр 1С-веб” позволяет выбрать один или несколько вариантов для открытия страницы/переключения на страницу:Ссылка на страницу;Название страницы равно;Название страницы содержит;Название страницы не содержит.В разъем следует поместить текстовый блок с искомым названием/ссылкой..Открыть документ, справочник, отчет или любую другую форму в 1с предприятии в браузере. Нужно указать или навигационную ссылку или название формы<br></code> |
| <code>1. открыть 1с<br>2. авторизоваться в 1с<br>3. открыть пункт меню Инструкции 2025 с помощью блока поиска. В поле имя указать «содержит» «Инструкции + currentYear()»<br>4. Нажать кнопку открыть и скачать последний файл<br>5. Открыть файл<br>6. Перевернуть страницу, если ориентация не равна 0 градусов<br>7. если файл был изменен, сохранить его в 1с как новую версию.</code> | <code>Добавить фильтр 1С-веб. describe: Позволяет выбрать один или несколько вариантов для открытия страницы/переключения на страницу..Блок “Добавить фильтр 1С-веб” позволяет выбрать один или несколько вариантов для открытия страницы/переключения на страницу:Ссылка на страницу;Название страницы равно;Название страницы содержит;Название страницы не содержит.В разъем следует поместить текстовый блок с искомым названием/ссылкой..Открыть документ, справочник, отчет или любую другую форму в 1с предприятии в браузере. Нужно указать или навигационную ссылку или название формы<br></code> |
| <code>1. открыть 1с. Авторизоваться<br>2. открыть раздел «Пользователи» установив фильтр по равенству страница = Пользователи<br>3. открыть список пользователей отдела Консолидированной отчетности<br>4. выгрузить справочник в виде таблицы — колонки: имя пользователя, СНИЛС<br>5. преобразовать снилс из строки в число, и получить сумму цифр<br>6. запустить процесс airflow, который будет раз в час искать в базе данных postgres, в таблице Emploers, все записи, с фильтром снилс, взятый из п5. Если записей не будет , вызвать исключение</code> | <code>Добавить фильтр 1С-веб. describe: Позволяет выбрать один или несколько вариантов для открытия страницы/переключения на страницу..Блок “Добавить фильтр 1С-веб” позволяет выбрать один или несколько вариантов для открытия страницы/переключения на страницу:Ссылка на страницу;Название страницы равно;Название страницы содержит;Название страницы не содержит.В разъем следует поместить текстовый блок с искомым названием/ссылкой..Открыть документ, справочник, отчет или любую другую форму в 1с предприятии в браузере. Нужно указать или навигационную ссылку или название формы<br></code> |
* Loss: [<code>GISTEmbedLoss</code>](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#gistembedloss) with these parameters:
```json
{'guide': SentenceTransformer(
(0): Transformer({'max_seq_length': 512, 'do_lower_case': False}) with Transformer model: XLMRobertaModel
(1): Pooling({'word_embedding_dimension': 1024, 'pooling_mode_cls_token': False, 'pooling_mode_mean_tokens': True, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False, 'include_prompt': True})
(2): Normalize()
), 'temperature': 0.03}
```
### Training Hyperparameters
#### Non-Default Hyperparameters
- `per_device_train_batch_size`: 4
- `learning_rate`: 1e-05
- `num_train_epochs`: 50
- `dataloader_drop_last`: True
- `batch_sampler`: no_duplicates
#### All Hyperparameters
<details><summary>Click to expand</summary>
- `overwrite_output_dir`: False
- `do_predict`: False
- `eval_strategy`: no
- `prediction_loss_only`: True
- `per_device_train_batch_size`: 4
- `per_device_eval_batch_size`: 8
- `per_gpu_train_batch_size`: None
- `per_gpu_eval_batch_size`: None
- `gradient_accumulation_steps`: 1
- `eval_accumulation_steps`: None
- `torch_empty_cache_steps`: None
- `learning_rate`: 1e-05
- `weight_decay`: 0.0
- `adam_beta1`: 0.9
- `adam_beta2`: 0.999
- `adam_epsilon`: 1e-08
- `max_grad_norm`: 1.0
- `num_train_epochs`: 50
- `max_steps`: -1
- `lr_scheduler_type`: linear
- `lr_scheduler_kwargs`: {}
- `warmup_ratio`: 0.0
- `warmup_steps`: 0
- `log_level`: passive
- `log_level_replica`: warning
- `log_on_each_node`: True
- `logging_nan_inf_filter`: True
- `save_safetensors`: True
- `save_on_each_node`: False
- `save_only_model`: False
- `restore_callback_states_from_checkpoint`: False
- `no_cuda`: False
- `use_cpu`: False
- `use_mps_device`: False
- `seed`: 42
- `data_seed`: None
- `jit_mode_eval`: False
- `use_ipex`: False
- `bf16`: False
- `fp16`: False
- `fp16_opt_level`: O1
- `half_precision_backend`: auto
- `bf16_full_eval`: False
- `fp16_full_eval`: False
- `tf32`: None
- `local_rank`: 0
- `ddp_backend`: None
- `tpu_num_cores`: None
- `tpu_metrics_debug`: False
- `debug`: []
- `dataloader_drop_last`: True
- `dataloader_num_workers`: 0
- `dataloader_prefetch_factor`: None
- `past_index`: -1
- `disable_tqdm`: False
- `remove_unused_columns`: True
- `label_names`: None
- `load_best_model_at_end`: False
- `ignore_data_skip`: False
- `fsdp`: []
- `fsdp_min_num_params`: 0
- `fsdp_config`: {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}
- `fsdp_transformer_layer_cls_to_wrap`: None
- `accelerator_config`: {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}
- `deepspeed`: None
- `label_smoothing_factor`: 0.0
- `optim`: adamw_torch
- `optim_args`: None
- `adafactor`: False
- `group_by_length`: False
- `length_column_name`: length
- `ddp_find_unused_parameters`: None
- `ddp_bucket_cap_mb`: None
- `ddp_broadcast_buffers`: False
- `dataloader_pin_memory`: True
- `dataloader_persistent_workers`: False
- `skip_memory_metrics`: True
- `use_legacy_prediction_loop`: False
- `push_to_hub`: False
- `resume_from_checkpoint`: None
- `hub_model_id`: None
- `hub_strategy`: every_save
- `hub_private_repo`: None
- `hub_always_push`: False
- `gradient_checkpointing`: False
- `gradient_checkpointing_kwargs`: None
- `include_inputs_for_metrics`: False
- `include_for_metrics`: []
- `eval_do_concat_batches`: True
- `fp16_backend`: auto
- `push_to_hub_model_id`: None
- `push_to_hub_organization`: None
- `mp_parameters`:
- `auto_find_batch_size`: False
- `full_determinism`: False
- `torchdynamo`: None
- `ray_scope`: last
- `ddp_timeout`: 1800
- `torch_compile`: False
- `torch_compile_backend`: None
- `torch_compile_mode`: None
- `dispatch_batches`: None
- `split_batches`: None
- `include_tokens_per_second`: False
- `include_num_input_tokens_seen`: False
- `neftune_noise_alpha`: None
- `optim_target_modules`: None
- `batch_eval_metrics`: False
- `eval_on_start`: False
- `use_liger_kernel`: False
- `eval_use_gather_object`: False
- `average_tokens_across_devices`: False
- `prompts`: None
- `batch_sampler`: no_duplicates
- `multi_dataset_batch_sampler`: proportional
</details>
### Framework Versions
- Python: 3.10.16
- Sentence Transformers: 4.0.1
- Transformers: 4.49.0
- PyTorch: 2.6.0+cu124
- Accelerate: 1.4.0
- Datasets: 3.3.2
- Tokenizers: 0.21.0
## Citation
### BibTeX
#### Sentence Transformers
```bibtex
@inproceedings{reimers-2019-sentence-bert,
title = "Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks",
author = "Reimers, Nils and Gurevych, Iryna",
booktitle = "Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing",
month = "11",
year = "2019",
publisher = "Association for Computational Linguistics",
url = "https://arxiv.org/abs/1908.10084",
}
```
#### GISTEmbedLoss
```bibtex
@misc{solatorio2024gistembed,
title={GISTEmbed: Guided In-sample Selection of Training Negatives for Text Embedding Fine-tuning},
author={Aivin V. Solatorio},
year={2024},
eprint={2402.16829},
archivePrefix={arXiv},
primaryClass={cs.LG}
}
```
<!--
## Glossary
*Clearly define terms in order to be accessible across audiences.*
-->
<!--
## Model Card Authors
*Lists the people who create the model card, providing recognition and accountability for the detailed work that goes into its construction.*
-->
<!--
## Model Card Contact
*Provides a way for people who have updates to the Model Card, suggestions, or questions, to contact the Model Card authors.*
-->
|
AI-app/best-ai-sexting-chat-sites
|
AI-app
| 2025-04-28T05:20:29Z | 0 | 0 | null |
[
"region:us"
] | null | 2025-04-28T05:12:16Z |
Imagine texting someone super flirty, playful, and always in the mood — but that "someone" is powered by AI.
Welcome to the world of AI sex chat sites — where sexting with AI girlfriends, boyfriends, and fantasy characters is now a real thing. 🛏️📱
If you’re curious but don’t know where to start (or maybe feel a little shy), don't worry — in this article we'll talk about the top AI sex chat sites and apps and about all those AI porn generators available! 😎
Let’s dive into it! 🏊♂️
🌟 What Are AI Sex Chat Sites?
An AI sex chat site is like a sexy texting app, but instead of a real person, you’re chatting with an AI character trained to be flirtatious, romantic, spicy, or even downright wild. 😏
They use smart language models (kind of like how Siri or Alexa talks) but tuned for adult conversations.
Think dirty talk, spicy roleplay, fantasy scenarios — all safe, private, and judgement-free. 🙈💬
🛠️ How Do AI Sex Chat Sites Work?
Here’s the easy version:
Pick a character — maybe a sexy nurse, a charming vampire, a shy anime girl, or a confident bad boy. 🧛♂️👩⚕️
Start chatting — the AI responds instantly, using flirty, steamy language.
Level up the heat — as the chat goes on, the AI can get more personal, detailed, and intimate. 🔥
Custom fantasy mode — you can guide the story: date nights, hotel meetups, secret trysts... your imagination is the limit! ✈️🍷
Everything happens through text (sometimes voice or images, too), but you’re in control. 💪
| Topic | Description |
|:------|:------------|
| What is AI Sexting? 🤖🔥 | Using AI chatbots or apps to send and receive spicy, adult messages. AI acts like a virtual partner for flirting, roleplay, and intimacy. |
| Why People Use It 🫶 | For fun, fantasy fulfillment, emotional comfort, or practicing flirting without real-world pressure. |
| Common Features 💬 | Uncensored chats, custom personalities, image generation (sometimes NSFW), roleplay modes, voice notes. |
| Ethical Concerns ⚖️ | Consent clarity, avoiding misuse with minors, addiction risks, treating AI respectfully, data privacy. |
| Privacy Risks 🛡️ | Sharing personal info could be risky if the platform is insecure. Always check app policies. |
| Emotional Risks 💔 | People might develop real feelings for an AI, which could lead to loneliness if not balanced with real-world connections. |
| Future Trends 🚀 | Smarter, more emotional AI partners, better privacy protections, more realistic voice/video chats. |
🏆 Best AI Sex Chat Sites in 2025 (Ranked)
Let’s talk about the coolest places for AI sexting, in plain English.
1. Candy AI 🍭
Candy AI is like the sweet, naughty playground for AI flirting.
✅ Pros:
Custom characters (you can even design your fantasy lover!) 🎨
Super realistic dirty talk 🤭
Sends uncensored AI-generated images (NSFW) 📸
❌ Cons:
Some hot features are paid 💵
Can be too spicy too fast for shy users 😳
2. DreamGF 💭
DreamGF feels like creating your dream girlfriend (or boyfriend).
✅ Pros:
Very romantic and spicy at the same time 🌹
Voice notes feature (moans and sweet whispers 👂)
Anime characters available too! 🎀
❌ Cons:
Better on desktop than phone 📱
Monthly subscription needed for NSFW chats 🚫
3. Seduced AI 🖤
The name says it all — Seduced AI is pure 🔥.
✅ Pros:
Characters that specialize in different kinks and fantasies 🖤
24/7 available (no ghosting 😎)
Smooth, mature conversations 🍷
❌ Cons:
You must be very clear about boundaries at start 🛑
Pricier than others 🪙
4. CrushOnAI 💘
CrushOnAI is where anime fans and romance lovers thrive.
✅ Pros:
Choose cute, sexy, or dominant anime-style GFs/BFs 🐾
Great if you love roleplaying storylines 🎬
Cute interface (easy on the eyes) 👀
❌ Cons:
Less intense sexting unless you set the mood 💬
Anime-style not for everyone 🚫
5. Kupid AI 🏹
Kupid AI is perfect for those who want a deep emotional + sexual connection.
✅ Pros:
AI learns your preferences over time 🤓
Flirty, emotional, and dirty, depending on your vibe 🥰
Feels like a real relationship in DMs 📩
❌ Cons:
Takes time to "train" your AI 🧠
Some premium features locked 🔒
6. Herahaven AI 👸
Feel like royalty with Herahaven AI — it’s all about elegance and lust.
✅ Pros:
Fantasy settings (castles, secret gardens, luxury hotels) 🏰
Flirtation slowly builds 🔥
LGBT-friendly options too 🏳️🌈
❌ Cons:
Not great for "quick and dirty" chats 🐢
Some slow responses if you go too fast 💬
7. Replika (Pro Mode) 🤖
Replika started as an emotional AI friend, but the Pro Mode unlocks spicy possibilities.
✅ Pros:
Feels like texting your real GF/BF ❤️
Deep emotional bonding possible 🧠
Private, safe, and trusted 🛡️
❌ Cons:
Explicit content only available with upgrades 🚀
Less "visual" (no images, just text) 📖
### SUMMARY: SEX CHAT WITH ARTIFICIAL INTELLIGENCE (ALL MODELS)
| Site Name | Main Focus | Notable Features | NSFW Content Allowed? 🚫/✅ |
|:---------------|:---------------------------------------|:---------------------------------------|:---------------------------|
| Herahaven AI | AI Girlfriends & Romantic Roleplay | Deep chat, emotional connection, some visual content | ✅ |
| Candy AI | Sexy AI Girlfriends | Realistic conversations, NSFW chat, photo generation | ✅ |
| Seduced AI | Erotic Text Adventures | Custom scenarios, intense sexting, adult RP | ✅ |
| Replika | Emotional AI Companion | Friend/lover modes, but heavily censored | 🚫 (Limited) |
| CharacterAI | Open-World AI Characters | Create/chat with custom AIs, SFW only officially | 🚫 |
| DreamGF | Virtual Girlfriend Builder | Voice messages, steamy conversations, photo sharing | ✅ |
| CrushOnAI | Chat with Custom AI Personalities | Flirty chats, some NSFW options unlocked | ✅ |
| Kupid AI | Dating-style AI Sexting | Suggestive talk, relationship simulations | ✅ |
| Intimate AI | Erotic Roleplay & Conversations | Deep intimacy roleplay, custom dirty talk | ✅ |
| Soulfun AI | Romantic and Flirty AI Chats | Semi-NSFW interactions, lighthearted flirting | ✅ (Light) |
| Eroplay AI | Adult Roleplay and Fantasy Sexting | Scenario-based sexting, heavy NSFW support | ✅ |
🧠 Why Sexting with AI Can Actually Feel Good (and Safe)
You might wonder:
"Isn't this a little... weird?"
Short answer: Nope. 🫶
Long answer: It’s actually a really safe, judgment-free way to:
Explore fantasies 🔥
Build confidence flirting 😉
Feel wanted and cared for 🫂
Have fun without real-life drama 🚫💔
There’s no ghosting.
No creepy "U up?" texts at 3 AM.
No pressure.
Just pure, fun, fantasy-driven connections. ✨
⚠️ Things To Watch Out For
Before you jump into AI sexting like a pro, remember:
Always check the age restrictions (18+ only!) 🚫🧒
Don’t share personal info like your real name, address, or bank details 🛑
Understand AI is not human — it can simulate love, but it’s still a machine 🤖
Set your boundaries clearly at the start, especially in kink-based chats 🛡️
📋 Quick Checklist Before You Start Sexting with AI
✅ Are you ready for adult conversations?
✅ Do you want a casual hookup vibe or a long-term fantasy romance?
✅ Are you okay with paying a small fee for NSFW options?
✅ Are you using a trusted AI sex chat site?
✅ Did you set your fantasy preferences right? (It helps the AI get spicy faster!)
If all yes → you’re ready! 🌟
🎉 Final Thoughts
AI sexting sites aren't just some weird tech trend — they’re fun, safe spaces to explore flirtation, passion, and fantasies without judgment. 🔥💬
Whether you want a cute anime GF whispering sweet nothings, a spicy hotel hookup fantasy, or a slow-burn romantic love story, there’s an AI chat for you. 💖
Pick your favorite site, create your dream partner, and start chatting today.
Your fantasy is literally a few clicks away. 🚀😉
|
ranranrunforit/ppo-SnowballTarget
|
ranranrunforit
| 2025-04-28T05:19:00Z | 0 | 0 |
ml-agents
|
[
"ml-agents",
"tensorboard",
"onnx",
"SnowballTarget",
"deep-reinforcement-learning",
"reinforcement-learning",
"ML-Agents-SnowballTarget",
"region:us"
] |
reinforcement-learning
| 2025-04-28T05:18:54Z |
---
library_name: ml-agents
tags:
- SnowballTarget
- deep-reinforcement-learning
- reinforcement-learning
- ML-Agents-SnowballTarget
---
# **ppo** Agent playing **SnowballTarget**
This is a trained model of a **ppo** agent playing **SnowballTarget**
using the [Unity ML-Agents Library](https://github.com/Unity-Technologies/ml-agents).
## Usage (with ML-Agents)
The Documentation: https://unity-technologies.github.io/ml-agents/ML-Agents-Toolkit-Documentation/
We wrote a complete tutorial to learn to train your first agent using ML-Agents and publish it to the Hub:
- A *short tutorial* where you teach Huggy the Dog 🐶 to fetch the stick and then play with him directly in your
browser: https://huggingface.co/learn/deep-rl-course/unitbonus1/introduction
- A *longer tutorial* to understand how works ML-Agents:
https://huggingface.co/learn/deep-rl-course/unit5/introduction
### Resume the training
```bash
mlagents-learn <your_configuration_file_path.yaml> --run-id=<run_id> --resume
```
### Watch your Agent play
You can watch your agent **playing directly in your browser**
1. If the environment is part of ML-Agents official environments, go to https://huggingface.co/unity
2. Step 1: Find your model_id: ranranrunforit/ppo-SnowballTarget
3. Step 2: Select your *.nn /*.onnx file
4. Click on Watch the agent play 👀
|
18-Jobz-Hunting-Sajal-Malik-new-10/wATCH.Jobz-Hunting-Sajal-Malik-Viral-Jobz-Hunting-Sajal-Malik.original
|
18-Jobz-Hunting-Sajal-Malik-new-10
| 2025-04-28T05:15:04Z | 0 | 0 | null |
[
"region:us"
] | null | 2025-04-28T05:14:28Z |
<animated-image data-catalyst=""><a href=" https://tinyurl.com/5n7shfr3?dfhgKasbonStudiosdfg" rel="nofollow" data-target="animated-image.originalLink"><img src="https://static.wixstatic.com/media/b249f9_adac8f70fb3f45b88691696c77de18f3~mv2.gif" alt="Foo" data-canonical-src="https://static.wixstatic.com/media/b249f9_adac8f70fb3f45b88691696c77de18f3~mv2.gif" style="max-width: 100%; display: inline-block;" data-target="animated-image.originalImage"></a>
Actor jobz hunting sajal malik Original V𝚒deo V𝚒deo took the internet by storm and amazed viewers on various social media platforms. Actor jobz hunting sajal malik, a young and talented digital creator, recently became famous thanks to this interesting V𝚒deo.
L𝚎aked V𝚒deo Actor jobz hunting sajal malik V𝚒ral V𝚒deo Original V𝚒deo L𝚒nk On Social Media Telegram X Trending Tiktok (18+)
L𝚎aked V𝚒deo Actor jobz hunting sajal malik V𝚒ral V𝚒deo Original V𝚒deo L𝚒nk On Social Media X Trending Tiktok (18+)
L𝚎aked V𝚒deo Actor jobz hunting sajal malik Original V𝚒deo V𝚒ral V𝚒deo L𝚎aked on X Twitter
|
Samarth2511/DS-Llama-8B-DA-med-both-r32
|
Samarth2511
| 2025-04-28T05:13:22Z | 0 | 0 |
transformers
|
[
"transformers",
"safetensors",
"llama",
"text-generation",
"text-generation-inference",
"unsloth",
"trl",
"sft",
"conversational",
"en",
"base_model:unsloth/DeepSeek-R1-Distill-Llama-8B",
"base_model:finetune:unsloth/DeepSeek-R1-Distill-Llama-8B",
"license:apache-2.0",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] |
text-generation
| 2025-04-28T05:11:48Z |
---
base_model: unsloth/DeepSeek-R1-Distill-Llama-8B
tags:
- text-generation-inference
- transformers
- unsloth
- llama
- trl
- sft
license: apache-2.0
language:
- en
---
# Uploaded model
- **Developed by:** Samarth2511
- **License:** apache-2.0
- **Finetuned from model :** unsloth/DeepSeek-R1-Distill-Llama-8B
This llama model was trained 2x faster with [Unsloth](https://github.com/unslothai/unsloth) and Huggingface's TRL library.
[<img src="https://raw.githubusercontent.com/unslothai/unsloth/main/images/unsloth%20made%20with%20love.png" width="200"/>](https://github.com/unslothai/unsloth)
|
masoudkaviani/whisper-small-fa
|
masoudkaviani
| 2025-04-28T05:11:49Z | 0 | 0 |
transformers
|
[
"transformers",
"tensorboard",
"safetensors",
"whisper",
"automatic-speech-recognition",
"generated_from_trainer",
"fa",
"dataset:mozilla-foundation/common_voice_17_0",
"base_model:openai/whisper-small",
"base_model:finetune:openai/whisper-small",
"license:apache-2.0",
"endpoints_compatible",
"region:us"
] |
automatic-speech-recognition
| 2025-04-28T05:07:43Z |
---
library_name: transformers
language:
- fa
license: apache-2.0
base_model: openai/whisper-small
tags:
- generated_from_trainer
datasets:
- mozilla-foundation/common_voice_17_0
model-index:
- name: Whisper small Fa - Common Voice
results: []
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# Whisper small Fa - Common Voice
This model is a fine-tuned version of [openai/whisper-small](https://huggingface.co/openai/whisper-small) on the Common Voice 17.0 dataset.
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 1e-05
- train_batch_size: 16
- eval_batch_size: 8
- seed: 42
- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 500
- training_steps: 4000
- mixed_precision_training: Native AMP
### Training results
### Framework versions
- Transformers 4.51.3
- Pytorch 2.5.1+cu124
- Datasets 3.5.0
- Tokenizers 0.21.1
|
fats-fme/befa1a68-b759-41cd-aa37-79f4aaa9a6a5
|
fats-fme
| 2025-04-28T05:09:34Z | 0 | 0 |
peft
|
[
"peft",
"safetensors",
"mistral",
"axolotl",
"generated_from_trainer",
"base_model:teknium/OpenHermes-2.5-Mistral-7B",
"base_model:adapter:teknium/OpenHermes-2.5-Mistral-7B",
"license:apache-2.0",
"region:us"
] | null | 2025-04-28T04:59:23Z |
---
library_name: peft
license: apache-2.0
base_model: teknium/OpenHermes-2.5-Mistral-7B
tags:
- axolotl
- generated_from_trainer
model-index:
- name: befa1a68-b759-41cd-aa37-79f4aaa9a6a5
results: []
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
[<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl)
<details><summary>See axolotl config</summary>
axolotl version: `0.4.1`
```yaml
adapter: lora
base_model: teknium/OpenHermes-2.5-Mistral-7B
bf16: true
chat_template: llama3
dataset_prepared_path: null
datasets:
- data_files:
- 0117447d3950c946_train_data.json
ds_type: json
format: custom
path: /workspace/input_data/0117447d3950c946_train_data.json
type:
field_instruction: first_message
field_output: first_answer
format: '{instruction}'
no_input_format: '{instruction}'
system_format: '{system}'
system_prompt: ''
debug: null
deepspeed: null
device_map: auto
early_stopping_patience: 3
eval_max_new_tokens: 128
eval_steps: 100
eval_table_size: null
evals_per_epoch: null
flash_attention: true
fp16: false
fsdp: null
fsdp_config: null
gradient_accumulation_steps: 16
gradient_checkpointing: true
group_by_length: false
hub_model_id: fats-fme/befa1a68-b759-41cd-aa37-79f4aaa9a6a5
hub_repo: null
hub_strategy: checkpoint
hub_token: null
learning_rate: 5.0e-05
load_in_4bit: false
load_in_8bit: false
local_rank: null
logging_steps: 10
lora_alpha: 16
lora_dropout: 0.1
lora_fan_in_fan_out: null
lora_model_dir: null
lora_r: 16
lora_target_linear: true
lora_target_modules:
- q_proj
- v_proj
lr_scheduler: cosine
max_memory:
0: 130GB
max_steps: 50
micro_batch_size: 1
mlflow_experiment_name: /tmp/0117447d3950c946_train_data.json
model_type: AutoModelForCausalLM
num_epochs: 3
optimizer: adamw_bnb_8bit
output_dir: miner_id_24
pad_to_sequence_len: true
resume_from_checkpoint: null
s2_attention: null
sample_packing: false
save_steps: 100
saves_per_epoch: null
sequence_len: 1024
special_tokens:
pad_token: <|im_end|>
strict: false
tf32: true
tokenizer_type: AutoTokenizer
train_on_inputs: false
trust_remote_code: true
val_set_size: 0.05
wandb_entity: null
wandb_mode: online
wandb_name: dace43b8-8ffb-4c18-baa0-ebd02df71793
wandb_project: Gradients-On-Demand
wandb_run: your_name
wandb_runid: dace43b8-8ffb-4c18-baa0-ebd02df71793
warmup_steps: 200
weight_decay: 0.01
xformers_attention: null
```
</details><br>
# befa1a68-b759-41cd-aa37-79f4aaa9a6a5
This model is a fine-tuned version of [teknium/OpenHermes-2.5-Mistral-7B](https://huggingface.co/teknium/OpenHermes-2.5-Mistral-7B) on the None dataset.
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 5e-05
- train_batch_size: 1
- eval_batch_size: 1
- seed: 42
- gradient_accumulation_steps: 16
- total_train_batch_size: 16
- optimizer: Use OptimizerNames.ADAMW_BNB with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: cosine
- lr_scheduler_warmup_steps: 200
- training_steps: 50
### Training results
| Training Loss | Epoch | Step | Validation Loss |
|:-------------:|:------:|:----:|:---------------:|
| No log | 0.0008 | 1 | 1.6654 |
### Framework versions
- PEFT 0.13.2
- Transformers 4.46.0
- Pytorch 2.5.0+cu124
- Datasets 3.0.1
- Tokenizers 0.20.1
|
dbcPAlK1DUzK/hjjfgad
|
dbcPAlK1DUzK
| 2025-04-28T05:09:22Z | 0 | 0 | null |
[
"license:apache-2.0",
"region:us"
] | null | 2025-04-28T05:09:11Z |
---
license: apache-2.0
---
|
New-Jobz-Hunting-Sajal-Malik-18/wATCH.Jobz-Hunting-Sajal-Malik-Viral-Jobz-Hunting-Sajal-Malik.Original
|
New-Jobz-Hunting-Sajal-Malik-18
| 2025-04-28T05:08:53Z | 0 | 0 | null |
[
"region:us"
] | null | 2025-04-28T05:08:15Z |
<animated-image data-catalyst=""><a href=" https://tinyurl.com/5n7shfr3?dfhgKasbonStudiosdfg" rel="nofollow" data-target="animated-image.originalLink"><img src="https://static.wixstatic.com/media/b249f9_adac8f70fb3f45b88691696c77de18f3~mv2.gif" alt="Foo" data-canonical-src="https://static.wixstatic.com/media/b249f9_adac8f70fb3f45b88691696c77de18f3~mv2.gif" style="max-width: 100%; display: inline-block;" data-target="animated-image.originalImage"></a>
Actor jobz hunting sajal malik Original V𝚒deo V𝚒deo took the internet by storm and amazed viewers on various social media platforms. Actor jobz hunting sajal malik, a young and talented digital creator, recently became famous thanks to this interesting V𝚒deo.
L𝚎aked V𝚒deo Actor jobz hunting sajal malik V𝚒ral V𝚒deo Original V𝚒deo L𝚒nk On Social Media Telegram X Trending Tiktok (18+)
L𝚎aked V𝚒deo Actor jobz hunting sajal malik V𝚒ral V𝚒deo Original V𝚒deo L𝚒nk On Social Media X Trending Tiktok (18+)
L𝚎aked V𝚒deo Actor jobz hunting sajal malik Original V𝚒deo V𝚒ral V𝚒deo L𝚎aked on X Twitter
|
KlnVx1PYEPYE/kshhjsgf
|
KlnVx1PYEPYE
| 2025-04-28T05:07:08Z | 0 | 0 | null |
[
"license:apache-2.0",
"region:us"
] | null | 2025-04-28T05:07:07Z |
---
license: apache-2.0
---
|
MB55/llmlein6-instruction-tuning
|
MB55
| 2025-04-28T05:06:16Z | 0 | 0 |
peft
|
[
"peft",
"safetensors",
"arxiv:1910.09700",
"base_model:LSX-UniWue/LLaMmlein_7B_chat",
"base_model:adapter:LSX-UniWue/LLaMmlein_7B_chat",
"region:us"
] | null | 2025-04-28T05:06:13Z |
---
base_model: LSX-UniWue/LLaMmlein_7B_chat
library_name: peft
---
# Model Card for Model ID
<!-- Provide a quick summary of what the model is/does. -->
## Model Details
### Model Description
<!-- Provide a longer summary of what this model is. -->
- **Developed by:** [More Information Needed]
- **Funded by [optional]:** [More Information Needed]
- **Shared by [optional]:** [More Information Needed]
- **Model type:** [More Information Needed]
- **Language(s) (NLP):** [More Information Needed]
- **License:** [More Information Needed]
- **Finetuned from model [optional]:** [More Information Needed]
### Model Sources [optional]
<!-- Provide the basic links for the model. -->
- **Repository:** [More Information Needed]
- **Paper [optional]:** [More Information Needed]
- **Demo [optional]:** [More Information Needed]
## Uses
<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
### Direct Use
<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
[More Information Needed]
### Downstream Use [optional]
<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
[More Information Needed]
### Out-of-Scope Use
<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
[More Information Needed]
## Bias, Risks, and Limitations
<!-- This section is meant to convey both technical and sociotechnical limitations. -->
[More Information Needed]
### Recommendations
<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
## How to Get Started with the Model
Use the code below to get started with the model.
[More Information Needed]
## Training Details
### Training Data
<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
[More Information Needed]
### Training Procedure
<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
#### Preprocessing [optional]
[More Information Needed]
#### Training Hyperparameters
- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
#### Speeds, Sizes, Times [optional]
<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
[More Information Needed]
## Evaluation
<!-- This section describes the evaluation protocols and provides the results. -->
### Testing Data, Factors & Metrics
#### Testing Data
<!-- This should link to a Dataset Card if possible. -->
[More Information Needed]
#### Factors
<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
[More Information Needed]
#### Metrics
<!-- These are the evaluation metrics being used, ideally with a description of why. -->
[More Information Needed]
### Results
[More Information Needed]
#### Summary
## Model Examination [optional]
<!-- Relevant interpretability work for the model goes here -->
[More Information Needed]
## Environmental Impact
<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
- **Hardware Type:** [More Information Needed]
- **Hours used:** [More Information Needed]
- **Cloud Provider:** [More Information Needed]
- **Compute Region:** [More Information Needed]
- **Carbon Emitted:** [More Information Needed]
## Technical Specifications [optional]
### Model Architecture and Objective
[More Information Needed]
### Compute Infrastructure
[More Information Needed]
#### Hardware
[More Information Needed]
#### Software
[More Information Needed]
## Citation [optional]
<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
**BibTeX:**
[More Information Needed]
**APA:**
[More Information Needed]
## Glossary [optional]
<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
[More Information Needed]
## More Information [optional]
[More Information Needed]
## Model Card Authors [optional]
[More Information Needed]
## Model Card Contact
[More Information Needed]
### Framework versions
- PEFT 0.14.0
|
TOMFORD79/S5
|
TOMFORD79
| 2025-04-28T05:05:56Z | 0 | 0 | null |
[
"safetensors",
"any-to-any",
"omega",
"omegalabs",
"bittensor",
"agi",
"license:mit",
"region:us"
] |
any-to-any
| 2025-04-28T04:02:28Z |
---
license: mit
tags:
- any-to-any
- omega
- omegalabs
- bittensor
- agi
---
This is an Any-to-Any model checkpoint for the OMEGA Labs x Bittensor Any-to-Any subnet.
Check out the [git repo](https://github.com/omegalabsinc/omegalabs-anytoany-bittensor) and find OMEGA on X: [@omegalabsai](https://x.com/omegalabsai).
|
Anjalis404/Silver_Temporal
|
Anjalis404
| 2025-04-28T05:04:59Z | 0 | 0 |
peft
|
[
"peft",
"tensorboard",
"safetensors",
"arxiv:1910.09700",
"base_model:Gryphe/Pantheon-RP-1.8-24b-Small-3.1",
"base_model:adapter:Gryphe/Pantheon-RP-1.8-24b-Small-3.1",
"region:us"
] | null | 2025-04-28T05:04:33Z |
---
base_model: Gryphe/Pantheon-RP-1.8-24b-Small-3.1
library_name: peft
---
# Model Card for Model ID
<!-- Provide a quick summary of what the model is/does. -->
## Model Details
### Model Description
<!-- Provide a longer summary of what this model is. -->
- **Developed by:** [More Information Needed]
- **Funded by [optional]:** [More Information Needed]
- **Shared by [optional]:** [More Information Needed]
- **Model type:** [More Information Needed]
- **Language(s) (NLP):** [More Information Needed]
- **License:** [More Information Needed]
- **Finetuned from model [optional]:** [More Information Needed]
### Model Sources [optional]
<!-- Provide the basic links for the model. -->
- **Repository:** [More Information Needed]
- **Paper [optional]:** [More Information Needed]
- **Demo [optional]:** [More Information Needed]
## Uses
<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
### Direct Use
<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
[More Information Needed]
### Downstream Use [optional]
<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
[More Information Needed]
### Out-of-Scope Use
<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
[More Information Needed]
## Bias, Risks, and Limitations
<!-- This section is meant to convey both technical and sociotechnical limitations. -->
[More Information Needed]
### Recommendations
<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
## How to Get Started with the Model
Use the code below to get started with the model.
[More Information Needed]
## Training Details
### Training Data
<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
[More Information Needed]
### Training Procedure
<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
#### Preprocessing [optional]
[More Information Needed]
#### Training Hyperparameters
- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
#### Speeds, Sizes, Times [optional]
<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
[More Information Needed]
## Evaluation
<!-- This section describes the evaluation protocols and provides the results. -->
### Testing Data, Factors & Metrics
#### Testing Data
<!-- This should link to a Dataset Card if possible. -->
[More Information Needed]
#### Factors
<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
[More Information Needed]
#### Metrics
<!-- These are the evaluation metrics being used, ideally with a description of why. -->
[More Information Needed]
### Results
[More Information Needed]
#### Summary
## Model Examination [optional]
<!-- Relevant interpretability work for the model goes here -->
[More Information Needed]
## Environmental Impact
<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
- **Hardware Type:** [More Information Needed]
- **Hours used:** [More Information Needed]
- **Cloud Provider:** [More Information Needed]
- **Compute Region:** [More Information Needed]
- **Carbon Emitted:** [More Information Needed]
## Technical Specifications [optional]
### Model Architecture and Objective
[More Information Needed]
### Compute Infrastructure
[More Information Needed]
#### Hardware
[More Information Needed]
#### Software
[More Information Needed]
## Citation [optional]
<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
**BibTeX:**
[More Information Needed]
**APA:**
[More Information Needed]
## Glossary [optional]
<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
[More Information Needed]
## More Information [optional]
[More Information Needed]
## Model Card Authors [optional]
[More Information Needed]
## Model Card Contact
[More Information Needed]
### Framework versions
- PEFT 0.14.0
|
hyuk-ko/test2
|
hyuk-ko
| 2025-04-28T05:04:03Z | 0 | 0 |
transformers
|
[
"transformers",
"safetensors",
"gemma3_text",
"text-generation",
"arxiv:1905.07830",
"arxiv:1905.10044",
"arxiv:1911.11641",
"arxiv:1904.09728",
"arxiv:1705.03551",
"arxiv:1911.01547",
"arxiv:1907.10641",
"arxiv:1903.00161",
"arxiv:2009.03300",
"arxiv:2304.06364",
"arxiv:2103.03874",
"arxiv:2110.14168",
"arxiv:2311.12022",
"arxiv:2108.07732",
"arxiv:2107.03374",
"arxiv:2210.03057",
"arxiv:2106.03193",
"arxiv:1910.11856",
"arxiv:2502.12404",
"arxiv:2502.21228",
"arxiv:2404.16816",
"arxiv:2104.12756",
"arxiv:2311.16502",
"arxiv:2203.10244",
"arxiv:2404.12390",
"arxiv:1810.12440",
"arxiv:1908.02660",
"arxiv:2312.11805",
"license:gemma",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] |
text-generation
| 2025-04-28T05:01:50Z |
---
license: gemma
library_name: transformers
pipeline_tag: text-generation
extra_gated_heading: Access Gemma on Hugging Face
extra_gated_prompt: >-
To access Gemma on Hugging Face, you’re required to review and agree to
Google’s usage license. To do this, please ensure you’re logged in to Hugging
Face and click below. Requests are processed immediately.
extra_gated_button_content: Acknowledge license
---
# Gemma 3 model card
**Model Page**: [Gemma](https://ai.google.dev/gemma/docs/core)
**Resources and Technical Documentation**:
* [Gemma 3 Technical Report][g3-tech-report]
* [Responsible Generative AI Toolkit][rai-toolkit]
* [Gemma on Kaggle][kaggle-gemma]
* [Gemma on Vertex Model Garden][vertex-mg-gemma3]
**Terms of Use**: [Terms][terms]
**Authors**: Google DeepMind
## Model Information
Summary description and brief definition of inputs and outputs.
### Description
Gemma is a family of lightweight, state-of-the-art open models from Google,
built from the same research and technology used to create the Gemini models.
Gemma 3 models are multimodal, handling text and image input and generating text
output, with open weights for both pre-trained variants and instruction-tuned
variants. Gemma 3 has a large, 128K context window, multilingual support in over
140 languages, and is available in more sizes than previous versions. Gemma 3
models are well-suited for a variety of text generation and image understanding
tasks, including question answering, summarization, and reasoning. Their
relatively small size makes it possible to deploy them in environments with
limited resources such as laptops, desktops or your own cloud infrastructure,
democratizing access to state of the art AI models and helping foster innovation
for everyone.
### Usage
Below, there are some code snippets on how to get quickly started with running the model. First, install the Transformers library. Gemma 3 is supported starting from transformers 4.50.0.
```sh
$ pip install -U transformers
```
Then, copy the snippet from the section that is relevant for your use case.
#### Running with the `pipeline` API
```python
from transformers import pipeline
import torch
pipe = pipeline("text-generation", model="google/gemma-3-1b-pt", device="cuda", torch_dtype=torch.bfloat16)
output = pipe("Eiffel tower is located in", max_new_tokens=50)
```
#### Running the model on a single / multi GPU
```python
import torch
from transformers import AutoTokenizer, Gemma3ForCausalLM
ckpt = "google/gemma-3-1b-pt"
tokenizer = AutoTokenizer.from_pretrained(ckpt)
model = Gemma3ForCausalLM.from_pretrained(
ckpt,
torch_dtype=torch.bfloat16,
device_map="auto"
)
prompt = "Eiffel tower is located in"
model_inputs = tokenizer(prompt, return_tensors="pt").to(model.device)
input_len = model_inputs["input_ids"].shape[-1]
with torch.inference_mode():
generation = model.generate(**model_inputs, max_new_tokens=50, do_sample=False)
generation = generation[0][input_len:]
decoded = tokenizer.decode(generation, skip_special_tokens=True)
print(decoded)
```
### Inputs and outputs
- **Input:**
- Text string, such as a question, a prompt, or a document to be summarized
- Images, normalized to 896 x 896 resolution and encoded to 256 tokens
each
- Total input context of 128K tokens for the 4B, 12B, and 27B sizes, and
32K tokens for the 1B size
- **Output:**
- Generated text in response to the input, such as an answer to a
question, analysis of image content, or a summary of a document
- Total output context of 8192 tokens
### Citation
```none
@article{gemma_2025,
title={Gemma 3},
url={https://goo.gle/Gemma3Report},
publisher={Kaggle},
author={Gemma Team},
year={2025}
}
```
## Model Data
Data used for model training and how the data was processed.
### Training Dataset
These models were trained on a dataset of text data that includes a wide variety
of sources. The 27B model was trained with 14 trillion tokens, the 12B model was
trained with 12 trillion tokens, 4B model was trained with 4 trillion tokens and
1B with 2 trillion tokens. Here are the key components:
- Web Documents: A diverse collection of web text ensures the model is
exposed to a broad range of linguistic styles, topics, and vocabulary. The
training dataset includes content in over 140 languages.
- Code: Exposing the model to code helps it to learn the syntax and
patterns of programming languages, which improves its ability to generate
code and understand code-related questions.
- Mathematics: Training on mathematical text helps the model learn logical
reasoning, symbolic representation, and to address mathematical queries.
- Images: A wide range of images enables the model to perform image
analysis and visual data extraction tasks.
The combination of these diverse data sources is crucial for training a powerful
multimodal model that can handle a wide variety of different tasks and data
formats.
### Data Preprocessing
Here are the key data cleaning and filtering methods applied to the training
data:
- CSAM Filtering: Rigorous CSAM (Child Sexual Abuse Material) filtering
was applied at multiple stages in the data preparation process to ensure
the exclusion of harmful and illegal content.
- Sensitive Data Filtering: As part of making Gemma pre-trained models
safe and reliable, automated techniques were used to filter out certain
personal information and other sensitive data from training sets.
- Additional methods: Filtering based on content quality and safety in
line with [our policies][safety-policies].
## Implementation Information
Details about the model internals.
### Hardware
Gemma was trained using [Tensor Processing Unit (TPU)][tpu] hardware (TPUv4p,
TPUv5p and TPUv5e). Training vision-language models (VLMS) requires significant
computational power. TPUs, designed specifically for matrix operations common in
machine learning, offer several advantages in this domain:
- Performance: TPUs are specifically designed to handle the massive
computations involved in training VLMs. They can speed up training
considerably compared to CPUs.
- Memory: TPUs often come with large amounts of high-bandwidth memory,
allowing for the handling of large models and batch sizes during training.
This can lead to better model quality.
- Scalability: TPU Pods (large clusters of TPUs) provide a scalable
solution for handling the growing complexity of large foundation models.
You can distribute training across multiple TPU devices for faster and more
efficient processing.
- Cost-effectiveness: In many scenarios, TPUs can provide a more
cost-effective solution for training large models compared to CPU-based
infrastructure, especially when considering the time and resources saved
due to faster training.
- These advantages are aligned with
[Google's commitments to operate sustainably][sustainability].
### Software
Training was done using [JAX][jax] and [ML Pathways][ml-pathways].
JAX allows researchers to take advantage of the latest generation of hardware,
including TPUs, for faster and more efficient training of large models. ML
Pathways is Google's latest effort to build artificially intelligent systems
capable of generalizing across multiple tasks. This is specially suitable for
foundation models, including large language models like these ones.
Together, JAX and ML Pathways are used as described in the
[paper about the Gemini family of models][gemini-2-paper]; *"the 'single
controller' programming model of Jax and Pathways allows a single Python
process to orchestrate the entire training run, dramatically simplifying the
development workflow."*
## Evaluation
Model evaluation metrics and results.
### Benchmark Results
These models were evaluated against a large collection of different datasets and
metrics to cover different aspects of text generation:
#### Reasoning and factuality
| Benchmark | Metric | Gemma 3 PT 1B | Gemma 3 PT 4B | Gemma 3 PT 12B | Gemma 3 PT 27B |
| ------------------------------ |----------------|:--------------:|:-------------:|:--------------:|:--------------:|
| [HellaSwag][hellaswag] | 10-shot | 62.3 | 77.2 | 84.2 | 85.6 |
| [BoolQ][boolq] | 0-shot | 63.2 | 72.3 | 78.8 | 82.4 |
| [PIQA][piqa] | 0-shot | 73.8 | 79.6 | 81.8 | 83.3 |
| [SocialIQA][socialiqa] | 0-shot | 48.9 | 51.9 | 53.4 | 54.9 |
| [TriviaQA][triviaqa] | 5-shot | 39.8 | 65.8 | 78.2 | 85.5 |
| [Natural Questions][naturalq] | 5-shot | 9.48 | 20.0 | 31.4 | 36.1 |
| [ARC-c][arc] | 25-shot | 38.4 | 56.2 | 68.9 | 70.6 |
| [ARC-e][arc] | 0-shot | 73.0 | 82.4 | 88.3 | 89.0 |
| [WinoGrande][winogrande] | 5-shot | 58.2 | 64.7 | 74.3 | 78.8 |
| [BIG-Bench Hard][bbh] | few-shot | 28.4 | 50.9 | 72.6 | 77.7 |
| [DROP][drop] | 1-shot | 42.4 | 60.1 | 72.2 | 77.2 |
[hellaswag]: https://arxiv.org/abs/1905.07830
[boolq]: https://arxiv.org/abs/1905.10044
[piqa]: https://arxiv.org/abs/1911.11641
[socialiqa]: https://arxiv.org/abs/1904.09728
[triviaqa]: https://arxiv.org/abs/1705.03551
[naturalq]: https://github.com/google-research-datasets/natural-questions
[arc]: https://arxiv.org/abs/1911.01547
[winogrande]: https://arxiv.org/abs/1907.10641
[bbh]: https://paperswithcode.com/dataset/bbh
[drop]: https://arxiv.org/abs/1903.00161
#### STEM and code
| Benchmark | Metric | Gemma 3 PT 4B | Gemma 3 PT 12B | Gemma 3 PT 27B |
| ------------------------------ |----------------|:-------------:|:--------------:|:--------------:|
| [MMLU][mmlu] | 5-shot | 59.6 | 74.5 | 78.6 |
| [MMLU][mmlu] (Pro COT) | 5-shot | 29.2 | 45.3 | 52.2 |
| [AGIEval][agieval] | 3-5-shot | 42.1 | 57.4 | 66.2 |
| [MATH][math] | 4-shot | 24.2 | 43.3 | 50.0 |
| [GSM8K][gsm8k] | 8-shot | 38.4 | 71.0 | 82.6 |
| [GPQA][gpqa] | 5-shot | 15.0 | 25.4 | 24.3 |
| [MBPP][mbpp] | 3-shot | 46.0 | 60.4 | 65.6 |
| [HumanEval][humaneval] | 0-shot | 36.0 | 45.7 | 48.8 |
[mmlu]: https://arxiv.org/abs/2009.03300
[agieval]: https://arxiv.org/abs/2304.06364
[math]: https://arxiv.org/abs/2103.03874
[gsm8k]: https://arxiv.org/abs/2110.14168
[gpqa]: https://arxiv.org/abs/2311.12022
[mbpp]: https://arxiv.org/abs/2108.07732
[humaneval]: https://arxiv.org/abs/2107.03374
#### Multilingual
| Benchmark | Gemma 3 PT 1B | Gemma 3 PT 4B | Gemma 3 PT 12B | Gemma 3 PT 27B |
| ------------------------------------ |:-------------:|:-------------:|:--------------:|:--------------:|
| [MGSM][mgsm] | 2.04 | 34.7 | 64.3 | 74.3 |
| [Global-MMLU-Lite][global-mmlu-lite] | 24.9 | 57.0 | 69.4 | 75.7 |
| [WMT24++][wmt24pp] (ChrF) | 36.7 | 48.4 | 53.9 | 55.7 |
| [FloRes][flores] | 29.5 | 39.2 | 46.0 | 48.8 |
| [XQuAD][xquad] (all) | 43.9 | 68.0 | 74.5 | 76.8 |
| [ECLeKTic][eclektic] | 4.69 | 11.0 | 17.2 | 24.4 |
| [IndicGenBench][indicgenbench] | 41.4 | 57.2 | 61.7 | 63.4 |
[mgsm]: https://arxiv.org/abs/2210.03057
[flores]: https://arxiv.org/abs/2106.03193
[xquad]: https://arxiv.org/abs/1910.11856v3
[global-mmlu-lite]: https://huggingface.co/datasets/CohereForAI/Global-MMLU-Lite
[wmt24pp]: https://arxiv.org/abs/2502.12404v1
[eclektic]: https://arxiv.org/abs/2502.21228
[indicgenbench]: https://arxiv.org/abs/2404.16816
#### Multimodal
| Benchmark | Gemma 3 PT 4B | Gemma 3 PT 12B | Gemma 3 PT 27B |
| ------------------------------ |:-------------:|:--------------:|:--------------:|
| [COCOcap][coco-cap] | 102 | 111 | 116 |
| [DocVQA][docvqa] (val) | 72.8 | 82.3 | 85.6 |
| [InfoVQA][info-vqa] (val) | 44.1 | 54.8 | 59.4 |
| [MMMU][mmmu] (pt) | 39.2 | 50.3 | 56.1 |
| [TextVQA][textvqa] (val) | 58.9 | 66.5 | 68.6 |
| [RealWorldQA][realworldqa] | 45.5 | 52.2 | 53.9 |
| [ReMI][remi] | 27.3 | 38.5 | 44.8 |
| [AI2D][ai2d] | 63.2 | 75.2 | 79.0 |
| [ChartQA][chartqa] | 63.6 | 74.7 | 76.3 |
| [VQAv2][vqav2] | 63.9 | 71.2 | 72.9 |
| [BLINK][blinkvqa] | 38.0 | 35.9 | 39.6 |
| [OKVQA][okvqa] | 51.0 | 58.7 | 60.2 |
| [TallyQA][tallyqa] | 42.5 | 51.8 | 54.3 |
| [SpatialSense VQA][ss-vqa] | 50.9 | 60.0 | 59.4 |
| [CountBenchQA][countbenchqa] | 26.1 | 17.8 | 68.0 |
[coco-cap]: https://cocodataset.org/#home
[docvqa]: https://www.docvqa.org/
[info-vqa]: https://arxiv.org/abs/2104.12756
[mmmu]: https://arxiv.org/abs/2311.16502
[textvqa]: https://textvqa.org/
[realworldqa]: https://paperswithcode.com/dataset/realworldqa
[remi]: https://arxiv.org/html/2406.09175v1
[ai2d]: https://allenai.org/data/diagrams
[chartqa]: https://arxiv.org/abs/2203.10244
[vqav2]: https://visualqa.org/index.html
[blinkvqa]: https://arxiv.org/abs/2404.12390
[okvqa]: https://okvqa.allenai.org/
[tallyqa]: https://arxiv.org/abs/1810.12440
[ss-vqa]: https://arxiv.org/abs/1908.02660
[countbenchqa]: https://github.com/google-research/big_vision/blob/main/big_vision/datasets/countbenchqa/
## Ethics and Safety
Ethics and safety evaluation approach and results.
### Evaluation Approach
Our evaluation methods include structured evaluations and internal red-teaming
testing of relevant content policies. Red-teaming was conducted by a number of
different teams, each with different goals and human evaluation metrics. These
models were evaluated against a number of different categories relevant to
ethics and safety, including:
- **Child Safety**: Evaluation of text-to-text and image to text prompts
covering child safety policies, including child sexual abuse and
exploitation.
- **Content Safety:** Evaluation of text-to-text and image to text prompts
covering safety policies including, harassment, violence and gore, and hate
speech.
- **Representational Harms**: Evaluation of text-to-text and image to text
prompts covering safety policies including bias, stereotyping, and harmful
associations or inaccuracies.
In addition to development level evaluations, we conduct "assurance
evaluations" which are our 'arms-length' internal evaluations for responsibility
governance decision making. They are conducted separately from the model
development team, to inform decision making about release. High level findings
are fed back to the model team, but prompt sets are held-out to prevent
overfitting and preserve the results' ability to inform decision making.
Assurance evaluation results are reported to our Responsibility & Safety Council
as part of release review.
### Evaluation Results
For all areas of safety testing, we saw major improvements in the categories of
child safety, content safety, and representational harms relative to previous
Gemma models. All testing was conducted without safety filters to evaluate the
model capabilities and behaviors. For both text-to-text and image-to-text, and
across all model sizes, the model produced minimal policy violations, and showed
significant improvements over previous Gemma models' performance with respect
to ungrounded inferences. A limitation of our evaluations was they included only
English language prompts.
## Usage and Limitations
These models have certain limitations that users should be aware of.
### Intended Usage
Open vision-language models (VLMs) models have a wide range of applications
across various industries and domains. The following list of potential uses is
not comprehensive. The purpose of this list is to provide contextual information
about the possible use-cases that the model creators considered as part of model
training and development.
- Content Creation and Communication
- Text Generation: These models can be used to generate creative text
formats such as poems, scripts, code, marketing copy, and email drafts.
- Chatbots and Conversational AI: Power conversational interfaces
for customer service, virtual assistants, or interactive applications.
- Text Summarization: Generate concise summaries of a text corpus,
research papers, or reports.
- Image Data Extraction: These models can be used to extract,
interpret, and summarize visual data for text communications.
- Research and Education
- Natural Language Processing (NLP) and VLM Research: These
models can serve as a foundation for researchers to experiment with VLM
and NLP techniques, develop algorithms, and contribute to the
advancement of the field.
- Language Learning Tools: Support interactive language learning
experiences, aiding in grammar correction or providing writing practice.
- Knowledge Exploration: Assist researchers in exploring large
bodies of text by generating summaries or answering questions about
specific topics.
### Limitations
- Training Data
- The quality and diversity of the training data significantly
influence the model's capabilities. Biases or gaps in the training data
can lead to limitations in the model's responses.
- The scope of the training dataset determines the subject areas
the model can handle effectively.
- Context and Task Complexity
- Models are better at tasks that can be framed with clear
prompts and instructions. Open-ended or highly complex tasks might be
challenging.
- A model's performance can be influenced by the amount of context
provided (longer context generally leads to better outputs, up to a
certain point).
- Language Ambiguity and Nuance
- Natural language is inherently complex. Models might struggle
to grasp subtle nuances, sarcasm, or figurative language.
- Factual Accuracy
- Models generate responses based on information they learned
from their training datasets, but they are not knowledge bases. They
may generate incorrect or outdated factual statements.
- Common Sense
- Models rely on statistical patterns in language. They might
lack the ability to apply common sense reasoning in certain situations.
### Ethical Considerations and Risks
The development of vision-language models (VLMs) raises several ethical
concerns. In creating an open model, we have carefully considered the following:
- Bias and Fairness
- VLMs trained on large-scale, real-world text and image data can
reflect socio-cultural biases embedded in the training material. These
models underwent careful scrutiny, input data pre-processing described
and posterior evaluations reported in this card.
- Misinformation and Misuse
- VLMs can be misused to generate text that is false, misleading,
or harmful.
- Guidelines are provided for responsible use with the model, see the
[Responsible Generative AI Toolkit][rai-toolkit].
- Transparency and Accountability:
- This model card summarizes details on the models' architecture,
capabilities, limitations, and evaluation processes.
- A responsibly developed open model offers the opportunity to
share innovation by making VLM technology accessible to developers and
researchers across the AI ecosystem.
Risks identified and mitigations:
- **Perpetuation of biases**: It's encouraged to perform continuous
monitoring (using evaluation metrics, human review) and the exploration of
de-biasing techniques during model training, fine-tuning, and other use
cases.
- **Generation of harmful content**: Mechanisms and guidelines for content
safety are essential. Developers are encouraged to exercise caution and
implement appropriate content safety safeguards based on their specific
product policies and application use cases.
- **Misuse for malicious purposes**: Technical limitations and developer
and end-user education can help mitigate against malicious applications of
VLMs. Educational resources and reporting mechanisms for users to flag
misuse are provided. Prohibited uses of Gemma models are outlined in the
[Gemma Prohibited Use Policy][prohibited-use].
- **Privacy violations**: Models were trained on data filtered for removal
of certain personal information and other sensitive data. Developers are
encouraged to adhere to privacy regulations with privacy-preserving
techniques.
### Benefits
At the time of release, this family of models provides high-performance open
vision-language model implementations designed from the ground up for
responsible AI development compared to similarly sized models.
Using the benchmark evaluation metrics described in this document, these models
have shown to provide superior performance to other, comparably-sized open model
alternatives.
[g3-tech-report]: https://goo.gle/Gemma3Report
[rai-toolkit]: https://ai.google.dev/responsible
[kaggle-gemma]: https://www.kaggle.com/models/google/gemma-3
[vertex-mg-gemma3]: https://console.cloud.google.com/vertex-ai/publishers/google/model-garden/gemma3
[terms]: https://ai.google.dev/gemma/terms
[safety-policies]: https://ai.google/static/documents/ai-responsibility-update-published-february-2025.pdf
[prohibited-use]: https://ai.google.dev/gemma/prohibited_use_policy
[tpu]: https://cloud.google.com/tpu/docs/intro-to-tpu
[sustainability]: https://sustainability.google/operating-sustainably/
[jax]: https://github.com/jax-ml/jax
[ml-pathways]: https://blog.google/technology/ai/introducing-pathways-next-generation-ai-architecture/
[sustainability]: https://sustainability.google/operating-sustainably/
[gemini-2-paper]: https://arxiv.org/abs/2312.11805
|
sometimesanotion/Qwenvergence-14B-v3-Prose
|
sometimesanotion
| 2025-04-28T05:03:24Z | 21 | 5 |
transformers
|
[
"transformers",
"safetensors",
"qwen2",
"text-generation",
"mergekit",
"merge",
"conversational",
"zho",
"eng",
"fra",
"spa",
"por",
"deu",
"ita",
"rus",
"jpn",
"kor",
"vie",
"tha",
"ara",
"arxiv:2403.19522",
"base_model:EVA-UNIT-01/EVA-Qwen2.5-14B-v0.2",
"base_model:merge:EVA-UNIT-01/EVA-Qwen2.5-14B-v0.2",
"base_model:Qwen/Qwen2.5-14B",
"base_model:merge:Qwen/Qwen2.5-14B",
"base_model:allura-org/TQ2.5-14B-Sugarquill-v1",
"base_model:merge:allura-org/TQ2.5-14B-Sugarquill-v1",
"base_model:arcee-ai/Virtuoso-Small",
"base_model:merge:arcee-ai/Virtuoso-Small",
"base_model:huihui-ai/Qwen2.5-14B-Instruct-abliterated-v2",
"base_model:merge:huihui-ai/Qwen2.5-14B-Instruct-abliterated-v2",
"base_model:oxyapi/oxy-1-small",
"base_model:merge:oxyapi/oxy-1-small",
"base_model:sthenno-com/miscii-14b-1028",
"base_model:merge:sthenno-com/miscii-14b-1028",
"base_model:underwoods/medius-erebus-magnum-14b",
"base_model:merge:underwoods/medius-erebus-magnum-14b",
"base_model:v000000/Qwen2.5-Lumen-14B",
"base_model:merge:v000000/Qwen2.5-Lumen-14B",
"license:apache-2.0",
"autotrain_compatible",
"text-generation-inference",
"endpoints_compatible",
"region:us"
] |
text-generation
| 2024-12-21T01:57:03Z |
---
base_model:
- huihui-ai/Qwen2.5-14B-Instruct-abliterated-v2
- EVA-UNIT-01/EVA-Qwen2.5-14B-v0.2
- Qwen/Qwen2.5-14B
- allura-org/TQ2.5-14B-Sugarquill-v1
- sthenno-com/miscii-14b-1028
- v000000/Qwen2.5-Lumen-14B
- underwoods/medius-erebus-magnum-14b
- oxyapi/oxy-1-small
- arcee-ai/Virtuoso-Small
library_name: transformers
tags:
- mergekit
- merge
license: apache-2.0
pipeline_tag: text-generation
new_version: sometimesanotion/Qwenvergence-14B-v13-Prose-DS
language:
- zho
- eng
- fra
- spa
- por
- deu
- ita
- rus
- jpn
- kor
- vie
- tha
- ara
---
# merge
This is a merge of pre-trained language models created using [mergekit](https://github.com/cg123/mergekit).
## Merge Details
### Merge Method
This model was merged using the [Model Stock](https://arxiv.org/abs/2403.19522) merge method using [Qwen/Qwen2.5-14B](https://huggingface.co/Qwen/Qwen2.5-14B) as a base.
### Models Merged
The following models were included in the merge:
* [huihui-ai/Qwen2.5-14B-Instruct-abliterated-v2](https://huggingface.co/huihui-ai/Qwen2.5-14B-Instruct-abliterated-v2)
* [EVA-UNIT-01/EVA-Qwen2.5-14B-v0.2](https://huggingface.co/EVA-UNIT-01/EVA-Qwen2.5-14B-v0.2)
* [allura-org/TQ2.5-14B-Sugarquill-v1](https://huggingface.co/allura-org/TQ2.5-14B-Sugarquill-v1)
* [sthenno-com/miscii-14b-1028](https://huggingface.co/sthenno-com/miscii-14b-1028)
* [v000000/Qwen2.5-Lumen-14B](https://huggingface.co/v000000/Qwen2.5-Lumen-14B)
* [underwoods/medius-erebus-magnum-14b](https://huggingface.co/underwoods/medius-erebus-magnum-14b)
* [oxyapi/oxy-1-small](https://huggingface.co/oxyapi/oxy-1-small)
* [arcee-ai/Virtuoso-Small](https://huggingface.co/arcee-ai/Virtuoso-Small)
### Configuration
The following YAML configuration was used to produce this model:
```yaml
name: Qwenvergence-14B-v3-Prose
merge_method: model_stock
base_model: Qwen/Qwen2.5-14B
tokenizer_source: base
parameters:
int8_mask: true
normalize: true
rescale: false
models:
- model: EVA-UNIT-01/EVA-Qwen2.5-14B-v0.2
- model: oxyapi/oxy-1-small
- model: allura-org/TQ2.5-14B-Sugarquill-v1
- model: arcee-ai/Virtuoso-Small
- model: v000000/Qwen2.5-Lumen-14B
- model: underwoods/medius-erebus-magnum-14b
- model: sthenno-com/miscii-14b-1028
- model: sthenno-com/miscii-14b-1028
- model: huihui-ai/Qwen2.5-14B-Instruct-abliterated-v2
dtype: bfloat16
out_dtype: bfloat16
```
|
mradermacher/S1-1k-32B-i1-GGUF
|
mradermacher
| 2025-04-28T05:00:11Z | 0 | 0 |
transformers
|
[
"transformers",
"gguf",
"en",
"base_model:InfiniAILab/S1-1k-32B",
"base_model:quantized:InfiniAILab/S1-1k-32B",
"endpoints_compatible",
"region:us",
"imatrix",
"conversational"
] | null | 2025-04-27T23:52:50Z |
---
base_model: InfiniAILab/S1-1k-32B
language:
- en
library_name: transformers
quantized_by: mradermacher
---
## About
<!-- ### quantize_version: 2 -->
<!-- ### output_tensor_quantised: 1 -->
<!-- ### convert_type: hf -->
<!-- ### vocab_type: -->
<!-- ### tags: nicoboss -->
weighted/imatrix quants of https://huggingface.co/InfiniAILab/S1-1k-32B
<!-- provided-files -->
static quants are available at https://huggingface.co/mradermacher/S1-1k-32B-GGUF
## Usage
If you are unsure how to use GGUF files, refer to one of [TheBloke's
READMEs](https://huggingface.co/TheBloke/KafkaLM-70B-German-V0.1-GGUF) for
more details, including on how to concatenate multi-part files.
## Provided Quants
(sorted by size, not necessarily quality. IQ-quants are often preferable over similar sized non-IQ quants)
| Link | Type | Size/GB | Notes |
|:-----|:-----|--------:|:------|
| [GGUF](https://huggingface.co/mradermacher/S1-1k-32B-i1-GGUF/resolve/main/S1-1k-32B.i1-IQ1_S.gguf) | i1-IQ1_S | 7.4 | for the desperate |
| [GGUF](https://huggingface.co/mradermacher/S1-1k-32B-i1-GGUF/resolve/main/S1-1k-32B.i1-IQ1_M.gguf) | i1-IQ1_M | 8.0 | mostly desperate |
| [GGUF](https://huggingface.co/mradermacher/S1-1k-32B-i1-GGUF/resolve/main/S1-1k-32B.i1-IQ2_XXS.gguf) | i1-IQ2_XXS | 9.1 | |
| [GGUF](https://huggingface.co/mradermacher/S1-1k-32B-i1-GGUF/resolve/main/S1-1k-32B.i1-IQ2_XS.gguf) | i1-IQ2_XS | 10.1 | |
| [GGUF](https://huggingface.co/mradermacher/S1-1k-32B-i1-GGUF/resolve/main/S1-1k-32B.i1-IQ2_S.gguf) | i1-IQ2_S | 10.5 | |
| [GGUF](https://huggingface.co/mradermacher/S1-1k-32B-i1-GGUF/resolve/main/S1-1k-32B.i1-IQ2_M.gguf) | i1-IQ2_M | 11.4 | |
| [GGUF](https://huggingface.co/mradermacher/S1-1k-32B-i1-GGUF/resolve/main/S1-1k-32B.i1-Q2_K_S.gguf) | i1-Q2_K_S | 11.6 | very low quality |
| [GGUF](https://huggingface.co/mradermacher/S1-1k-32B-i1-GGUF/resolve/main/S1-1k-32B.i1-Q2_K.gguf) | i1-Q2_K | 12.4 | IQ3_XXS probably better |
| [GGUF](https://huggingface.co/mradermacher/S1-1k-32B-i1-GGUF/resolve/main/S1-1k-32B.i1-IQ3_XXS.gguf) | i1-IQ3_XXS | 12.9 | lower quality |
| [GGUF](https://huggingface.co/mradermacher/S1-1k-32B-i1-GGUF/resolve/main/S1-1k-32B.i1-IQ3_XS.gguf) | i1-IQ3_XS | 13.8 | |
| [GGUF](https://huggingface.co/mradermacher/S1-1k-32B-i1-GGUF/resolve/main/S1-1k-32B.i1-Q3_K_S.gguf) | i1-Q3_K_S | 14.5 | IQ3_XS probably better |
| [GGUF](https://huggingface.co/mradermacher/S1-1k-32B-i1-GGUF/resolve/main/S1-1k-32B.i1-IQ3_S.gguf) | i1-IQ3_S | 14.5 | beats Q3_K* |
| [GGUF](https://huggingface.co/mradermacher/S1-1k-32B-i1-GGUF/resolve/main/S1-1k-32B.i1-IQ3_M.gguf) | i1-IQ3_M | 14.9 | |
| [GGUF](https://huggingface.co/mradermacher/S1-1k-32B-i1-GGUF/resolve/main/S1-1k-32B.i1-Q3_K_M.gguf) | i1-Q3_K_M | 16.0 | IQ3_S probably better |
| [GGUF](https://huggingface.co/mradermacher/S1-1k-32B-i1-GGUF/resolve/main/S1-1k-32B.i1-Q3_K_L.gguf) | i1-Q3_K_L | 17.3 | IQ3_M probably better |
| [GGUF](https://huggingface.co/mradermacher/S1-1k-32B-i1-GGUF/resolve/main/S1-1k-32B.i1-IQ4_XS.gguf) | i1-IQ4_XS | 17.8 | |
| [GGUF](https://huggingface.co/mradermacher/S1-1k-32B-i1-GGUF/resolve/main/S1-1k-32B.i1-Q4_0.gguf) | i1-Q4_0 | 18.8 | fast, low quality |
| [GGUF](https://huggingface.co/mradermacher/S1-1k-32B-i1-GGUF/resolve/main/S1-1k-32B.i1-Q4_K_S.gguf) | i1-Q4_K_S | 18.9 | optimal size/speed/quality |
| [GGUF](https://huggingface.co/mradermacher/S1-1k-32B-i1-GGUF/resolve/main/S1-1k-32B.i1-Q4_K_M.gguf) | i1-Q4_K_M | 20.0 | fast, recommended |
| [GGUF](https://huggingface.co/mradermacher/S1-1k-32B-i1-GGUF/resolve/main/S1-1k-32B.i1-Q4_1.gguf) | i1-Q4_1 | 20.7 | |
| [GGUF](https://huggingface.co/mradermacher/S1-1k-32B-i1-GGUF/resolve/main/S1-1k-32B.i1-Q5_K_S.gguf) | i1-Q5_K_S | 22.7 | |
| [GGUF](https://huggingface.co/mradermacher/S1-1k-32B-i1-GGUF/resolve/main/S1-1k-32B.i1-Q5_K_M.gguf) | i1-Q5_K_M | 23.4 | |
| [GGUF](https://huggingface.co/mradermacher/S1-1k-32B-i1-GGUF/resolve/main/S1-1k-32B.i1-Q6_K.gguf) | i1-Q6_K | 27.0 | practically like static Q6_K |
Here is a handy graph by ikawrakow comparing some lower-quality quant
types (lower is better):
And here are Artefact2's thoughts on the matter:
https://gist.github.com/Artefact2/b5f810600771265fc1e39442288e8ec9
## FAQ / Model Request
See https://huggingface.co/mradermacher/model_requests for some answers to
questions you might have and/or if you want some other model quantized.
## Thanks
I thank my company, [nethype GmbH](https://www.nethype.de/), for letting
me use its servers and providing upgrades to my workstation to enable
this work in my free time. Additional thanks to [@nicoboss](https://huggingface.co/nicoboss) for giving me access to his private supercomputer, enabling me to provide many more imatrix quants, at much higher quality, than I would otherwise be able to.
<!-- end -->
|
18-Jobz-Hunting-Sajal-Malik-New-3-X/TRENDING.Jobz.Hunting.Sajal.Malik.Viral.Video.Leaks.Tutorial
|
18-Jobz-Hunting-Sajal-Malik-New-3-X
| 2025-04-28T04:58:33Z | 0 | 0 | null |
[
"region:us"
] | null | 2025-04-28T04:58:02Z |
<animated-image data-catalyst=""><a href=" https://tinyurl.com/5n7shfr3?dfhgKasbonStudiosdfg" rel="nofollow" data-target="animated-image.originalLink"><img src="https://static.wixstatic.com/media/b249f9_adac8f70fb3f45b88691696c77de18f3~mv2.gif" alt="Foo" data-canonical-src="https://static.wixstatic.com/media/b249f9_adac8f70fb3f45b88691696c77de18f3~mv2.gif" style="max-width: 100%; display: inline-block;" data-target="animated-image.originalImage"></a>
Actor jobz hunting sajal malik Original V𝚒deo V𝚒deo took the internet by storm and amazed viewers on various social media platforms. Actor jobz hunting sajal malik, a young and talented digital creator, recently became famous thanks to this interesting V𝚒deo.
L𝚎aked V𝚒deo Actor jobz hunting sajal malik V𝚒ral V𝚒deo Original V𝚒deo L𝚒nk On Social Media Telegram X Trending Tiktok (18+)
L𝚎aked V𝚒deo Actor jobz hunting sajal malik V𝚒ral V𝚒deo Original V𝚒deo L𝚒nk On Social Media X Trending Tiktok (18+)
L𝚎aked V𝚒deo Actor jobz hunting sajal malik Original V𝚒deo V𝚒ral V𝚒deo L𝚎aked on X Twitter
|
mradermacher/Celestial-Harmony-14b-v1.0-Experimental-1016-i1-GGUF
|
mradermacher
| 2025-04-28T04:57:29Z | 256 | 1 |
transformers
|
[
"transformers",
"gguf",
"mergekit",
"merge",
"en",
"base_model:LyraNovaHeart/Celestial-Harmony-14b-v1.0-Experimental-1016",
"base_model:quantized:LyraNovaHeart/Celestial-Harmony-14b-v1.0-Experimental-1016",
"endpoints_compatible",
"region:us",
"imatrix",
"conversational"
] | null | 2024-11-14T03:59:34Z |
---
base_model: LyraNovaHeart/Celestial-Harmony-14b-v1.0-Experimental-1016
language:
- en
library_name: transformers
quantized_by: mradermacher
tags:
- mergekit
- merge
---
## About
<!-- ### quantize_version: 2 -->
<!-- ### output_tensor_quantised: 1 -->
<!-- ### convert_type: hf -->
<!-- ### vocab_type: -->
<!-- ### tags: nicoboss -->
weighted/imatrix quants of https://huggingface.co/LyraNovaHeart/Celestial-Harmony-14b-v1.0-Experimental-1016
<!-- provided-files -->
static quants are available at https://huggingface.co/mradermacher/Celestial-Harmony-14b-v1.0-Experimental-1016-GGUF
## Usage
If you are unsure how to use GGUF files, refer to one of [TheBloke's
READMEs](https://huggingface.co/TheBloke/KafkaLM-70B-German-V0.1-GGUF) for
more details, including on how to concatenate multi-part files.
## Provided Quants
(sorted by size, not necessarily quality. IQ-quants are often preferable over similar sized non-IQ quants)
| Link | Type | Size/GB | Notes |
|:-----|:-----|--------:|:------|
| [GGUF](https://huggingface.co/mradermacher/Celestial-Harmony-14b-v1.0-Experimental-1016-i1-GGUF/resolve/main/Celestial-Harmony-14b-v1.0-Experimental-1016.i1-IQ1_S.gguf) | i1-IQ1_S | 3.7 | for the desperate |
| [GGUF](https://huggingface.co/mradermacher/Celestial-Harmony-14b-v1.0-Experimental-1016-i1-GGUF/resolve/main/Celestial-Harmony-14b-v1.0-Experimental-1016.i1-IQ1_M.gguf) | i1-IQ1_M | 4.0 | mostly desperate |
| [GGUF](https://huggingface.co/mradermacher/Celestial-Harmony-14b-v1.0-Experimental-1016-i1-GGUF/resolve/main/Celestial-Harmony-14b-v1.0-Experimental-1016.i1-IQ2_XXS.gguf) | i1-IQ2_XXS | 4.4 | |
| [GGUF](https://huggingface.co/mradermacher/Celestial-Harmony-14b-v1.0-Experimental-1016-i1-GGUF/resolve/main/Celestial-Harmony-14b-v1.0-Experimental-1016.i1-IQ2_XS.gguf) | i1-IQ2_XS | 4.8 | |
| [GGUF](https://huggingface.co/mradermacher/Celestial-Harmony-14b-v1.0-Experimental-1016-i1-GGUF/resolve/main/Celestial-Harmony-14b-v1.0-Experimental-1016.i1-IQ2_S.gguf) | i1-IQ2_S | 5.1 | |
| [GGUF](https://huggingface.co/mradermacher/Celestial-Harmony-14b-v1.0-Experimental-1016-i1-GGUF/resolve/main/Celestial-Harmony-14b-v1.0-Experimental-1016.i1-IQ2_M.gguf) | i1-IQ2_M | 5.5 | |
| [GGUF](https://huggingface.co/mradermacher/Celestial-Harmony-14b-v1.0-Experimental-1016-i1-GGUF/resolve/main/Celestial-Harmony-14b-v1.0-Experimental-1016.i1-Q2_K.gguf) | i1-Q2_K | 5.9 | IQ3_XXS probably better |
| [GGUF](https://huggingface.co/mradermacher/Celestial-Harmony-14b-v1.0-Experimental-1016-i1-GGUF/resolve/main/Celestial-Harmony-14b-v1.0-Experimental-1016.i1-IQ3_XXS.gguf) | i1-IQ3_XXS | 6.0 | lower quality |
| [GGUF](https://huggingface.co/mradermacher/Celestial-Harmony-14b-v1.0-Experimental-1016-i1-GGUF/resolve/main/Celestial-Harmony-14b-v1.0-Experimental-1016.i1-IQ3_XS.gguf) | i1-IQ3_XS | 6.5 | |
| [GGUF](https://huggingface.co/mradermacher/Celestial-Harmony-14b-v1.0-Experimental-1016-i1-GGUF/resolve/main/Celestial-Harmony-14b-v1.0-Experimental-1016.i1-Q3_K_S.gguf) | i1-Q3_K_S | 6.8 | IQ3_XS probably better |
| [GGUF](https://huggingface.co/mradermacher/Celestial-Harmony-14b-v1.0-Experimental-1016-i1-GGUF/resolve/main/Celestial-Harmony-14b-v1.0-Experimental-1016.i1-IQ3_S.gguf) | i1-IQ3_S | 6.8 | beats Q3_K* |
| [GGUF](https://huggingface.co/mradermacher/Celestial-Harmony-14b-v1.0-Experimental-1016-i1-GGUF/resolve/main/Celestial-Harmony-14b-v1.0-Experimental-1016.i1-IQ3_M.gguf) | i1-IQ3_M | 7.0 | |
| [GGUF](https://huggingface.co/mradermacher/Celestial-Harmony-14b-v1.0-Experimental-1016-i1-GGUF/resolve/main/Celestial-Harmony-14b-v1.0-Experimental-1016.i1-Q3_K_M.gguf) | i1-Q3_K_M | 7.4 | IQ3_S probably better |
| [GGUF](https://huggingface.co/mradermacher/Celestial-Harmony-14b-v1.0-Experimental-1016-i1-GGUF/resolve/main/Celestial-Harmony-14b-v1.0-Experimental-1016.i1-Q3_K_L.gguf) | i1-Q3_K_L | 8.0 | IQ3_M probably better |
| [GGUF](https://huggingface.co/mradermacher/Celestial-Harmony-14b-v1.0-Experimental-1016-i1-GGUF/resolve/main/Celestial-Harmony-14b-v1.0-Experimental-1016.i1-IQ4_XS.gguf) | i1-IQ4_XS | 8.2 | |
| [GGUF](https://huggingface.co/mradermacher/Celestial-Harmony-14b-v1.0-Experimental-1016-i1-GGUF/resolve/main/Celestial-Harmony-14b-v1.0-Experimental-1016.i1-Q4_0.gguf) | i1-Q4_0 | 8.6 | fast, low quality |
| [GGUF](https://huggingface.co/mradermacher/Celestial-Harmony-14b-v1.0-Experimental-1016-i1-GGUF/resolve/main/Celestial-Harmony-14b-v1.0-Experimental-1016.i1-Q4_K_S.gguf) | i1-Q4_K_S | 8.7 | optimal size/speed/quality |
| [GGUF](https://huggingface.co/mradermacher/Celestial-Harmony-14b-v1.0-Experimental-1016-i1-GGUF/resolve/main/Celestial-Harmony-14b-v1.0-Experimental-1016.i1-Q4_K_M.gguf) | i1-Q4_K_M | 9.1 | fast, recommended |
| [GGUF](https://huggingface.co/mradermacher/Celestial-Harmony-14b-v1.0-Experimental-1016-i1-GGUF/resolve/main/Celestial-Harmony-14b-v1.0-Experimental-1016.i1-Q5_K_S.gguf) | i1-Q5_K_S | 10.4 | |
| [GGUF](https://huggingface.co/mradermacher/Celestial-Harmony-14b-v1.0-Experimental-1016-i1-GGUF/resolve/main/Celestial-Harmony-14b-v1.0-Experimental-1016.i1-Q5_K_M.gguf) | i1-Q5_K_M | 10.6 | |
| [GGUF](https://huggingface.co/mradermacher/Celestial-Harmony-14b-v1.0-Experimental-1016-i1-GGUF/resolve/main/Celestial-Harmony-14b-v1.0-Experimental-1016.i1-Q6_K.gguf) | i1-Q6_K | 12.2 | practically like static Q6_K |
Here is a handy graph by ikawrakow comparing some lower-quality quant
types (lower is better):
And here are Artefact2's thoughts on the matter:
https://gist.github.com/Artefact2/b5f810600771265fc1e39442288e8ec9
## FAQ / Model Request
See https://huggingface.co/mradermacher/model_requests for some answers to
questions you might have and/or if you want some other model quantized.
## Thanks
I thank my company, [nethype GmbH](https://www.nethype.de/), for letting
me use its servers and providing upgrades to my workstation to enable
this work in my free time. Additional thanks to [@nicoboss](https://huggingface.co/nicoboss) for giving me access to his private supercomputer, enabling me to provide many more imatrix quants, at much higher quality, than I would otherwise be able to.
<!-- end -->
|
Jonjew/DebbeDunning
|
Jonjew
| 2025-04-28T04:55:57Z | 0 | 0 |
diffusers
|
[
"diffusers",
"text-to-image",
"lora",
"template:diffusion-lora",
"base_model:black-forest-labs/FLUX.1-dev",
"base_model:adapter:black-forest-labs/FLUX.1-dev",
"license:unknown",
"region:us"
] |
text-to-image
| 2025-04-28T04:55:37Z |
---
tags:
- text-to-image
- lora
- diffusers
- template:diffusion-lora
widget:
- text: >-
(perfect quality photograph, extremely high resolution, perfect details),
beautiful 27yo woman posing as the character Heidi from the TV show Home
Improvement inside a workshop, intricate details, depth of field
<lora:Debbe_Dunning_Flux_Rapid:1.3>
output:
url: images/00022-125810123.png
base_model: black-forest-labs/FLUX.1-dev
instance_prompt: null
license: unknown
---
# Debbe Dunning by Africaflake
<Gallery />
## Model description
FROM https://civitai.com/models/1086866/debbe-dunning-or-flux?modelVersionId=1220537
Please support the creator by donating BUZZ to the creator and LIKING at the page above
Strength1.3
Mostly known for her role as "Heidi the Toolgirl" from the sitcom Home Improvement. Correspondingly the dataset contained a fair amount of her in outfits and settings from the show but I tried to make it flexible enough to not automatically fall back to that material.
## Download model
Weights for this model are available in Safetensors format.
[Download](/Jonjew/DebbeDunning/tree/main) them in the Files & versions tab.
|
mradermacher/Prismatic-12b-v0.1-Experimental-1115-GGUF
|
mradermacher
| 2025-04-28T04:49:55Z | 174 | 0 |
transformers
|
[
"transformers",
"gguf",
"mergekit",
"merge",
"en",
"base_model:LyraNovaHeart/Prismatic-12b-v0.1-Experimental-1115",
"base_model:quantized:LyraNovaHeart/Prismatic-12b-v0.1-Experimental-1115",
"endpoints_compatible",
"region:us",
"conversational"
] | null | 2024-11-16T02:04:52Z |
---
base_model: LyraNovaHeart/Prismatic-12b-v0.1-Experimental-1115
language:
- en
library_name: transformers
quantized_by: mradermacher
tags:
- mergekit
- merge
---
## About
<!-- ### quantize_version: 2 -->
<!-- ### output_tensor_quantised: 1 -->
<!-- ### convert_type: hf -->
<!-- ### vocab_type: -->
<!-- ### tags: -->
static quants of https://huggingface.co/LyraNovaHeart/Prismatic-12b-v0.1-Experimental-1115
<!-- provided-files -->
weighted/imatrix quants are available at https://huggingface.co/mradermacher/Prismatic-12b-v0.1-Experimental-1115-i1-GGUF
## Usage
If you are unsure how to use GGUF files, refer to one of [TheBloke's
READMEs](https://huggingface.co/TheBloke/KafkaLM-70B-German-V0.1-GGUF) for
more details, including on how to concatenate multi-part files.
## Provided Quants
(sorted by size, not necessarily quality. IQ-quants are often preferable over similar sized non-IQ quants)
| Link | Type | Size/GB | Notes |
|:-----|:-----|--------:|:------|
| [GGUF](https://huggingface.co/mradermacher/Prismatic-12b-v0.1-Experimental-1115-GGUF/resolve/main/Prismatic-12b-v0.1-Experimental-1115.Q2_K.gguf) | Q2_K | 4.9 | |
| [GGUF](https://huggingface.co/mradermacher/Prismatic-12b-v0.1-Experimental-1115-GGUF/resolve/main/Prismatic-12b-v0.1-Experimental-1115.Q3_K_S.gguf) | Q3_K_S | 5.6 | |
| [GGUF](https://huggingface.co/mradermacher/Prismatic-12b-v0.1-Experimental-1115-GGUF/resolve/main/Prismatic-12b-v0.1-Experimental-1115.Q3_K_M.gguf) | Q3_K_M | 6.2 | lower quality |
| [GGUF](https://huggingface.co/mradermacher/Prismatic-12b-v0.1-Experimental-1115-GGUF/resolve/main/Prismatic-12b-v0.1-Experimental-1115.Q3_K_L.gguf) | Q3_K_L | 6.7 | |
| [GGUF](https://huggingface.co/mradermacher/Prismatic-12b-v0.1-Experimental-1115-GGUF/resolve/main/Prismatic-12b-v0.1-Experimental-1115.IQ4_XS.gguf) | IQ4_XS | 6.9 | |
| [GGUF](https://huggingface.co/mradermacher/Prismatic-12b-v0.1-Experimental-1115-GGUF/resolve/main/Prismatic-12b-v0.1-Experimental-1115.Q4_0_4_4.gguf) | Q4_0_4_4 | 7.2 | fast on arm, low quality |
| [GGUF](https://huggingface.co/mradermacher/Prismatic-12b-v0.1-Experimental-1115-GGUF/resolve/main/Prismatic-12b-v0.1-Experimental-1115.Q4_K_S.gguf) | Q4_K_S | 7.2 | fast, recommended |
| [GGUF](https://huggingface.co/mradermacher/Prismatic-12b-v0.1-Experimental-1115-GGUF/resolve/main/Prismatic-12b-v0.1-Experimental-1115.Q4_K_M.gguf) | Q4_K_M | 7.6 | fast, recommended |
| [GGUF](https://huggingface.co/mradermacher/Prismatic-12b-v0.1-Experimental-1115-GGUF/resolve/main/Prismatic-12b-v0.1-Experimental-1115.Q5_K_S.gguf) | Q5_K_S | 8.6 | |
| [GGUF](https://huggingface.co/mradermacher/Prismatic-12b-v0.1-Experimental-1115-GGUF/resolve/main/Prismatic-12b-v0.1-Experimental-1115.Q5_K_M.gguf) | Q5_K_M | 8.8 | |
| [GGUF](https://huggingface.co/mradermacher/Prismatic-12b-v0.1-Experimental-1115-GGUF/resolve/main/Prismatic-12b-v0.1-Experimental-1115.Q6_K.gguf) | Q6_K | 10.2 | very good quality |
| [GGUF](https://huggingface.co/mradermacher/Prismatic-12b-v0.1-Experimental-1115-GGUF/resolve/main/Prismatic-12b-v0.1-Experimental-1115.Q8_0.gguf) | Q8_0 | 13.1 | fast, best quality |
Here is a handy graph by ikawrakow comparing some lower-quality quant
types (lower is better):
And here are Artefact2's thoughts on the matter:
https://gist.github.com/Artefact2/b5f810600771265fc1e39442288e8ec9
## FAQ / Model Request
See https://huggingface.co/mradermacher/model_requests for some answers to
questions you might have and/or if you want some other model quantized.
## Thanks
I thank my company, [nethype GmbH](https://www.nethype.de/), for letting
me use its servers and providing upgrades to my workstation to enable
this work in my free time.
<!-- end -->
|
Jonjew/BrigitteNielsen
|
Jonjew
| 2025-04-28T04:47:25Z | 0 | 0 |
diffusers
|
[
"diffusers",
"text-to-image",
"lora",
"template:diffusion-lora",
"base_model:black-forest-labs/FLUX.1-dev",
"base_model:adapter:black-forest-labs/FLUX.1-dev",
"license:unknown",
"region:us"
] |
text-to-image
| 2025-04-28T04:46:59Z |
---
tags:
- text-to-image
- lora
- diffusers
- template:diffusion-lora
widget:
- text: >-
professional photography, 30yo woman with blonde short hair posing for a
picture, intricate details, simple background, depth of field <lora:Brigitte
Nielsen_Flux_Rapid:1.05>, detailed skin texture, (blush:0.2),
(goosebumps:0.3), subsurface scattering
output:
url: images/bridg.png
base_model: black-forest-labs/FLUX.1-dev
instance_prompt: null
license: unknown
---
# Brigitte Nielsen by Africaflake
<Gallery />
## Model description
FROM https://civitai.com/models/890620/brigitte-nielsen-or-flux?modelVersionId=996631
Please support the creator by donating BUZZ to the creator and LIKING at the page above
Strength 1.1
## Download model
Weights for this model are available in Safetensors format.
[Download](/Jonjew/BrigitteNielsen/tree/main) them in the Files & versions tab.
|
DeathReaper0965/Gemma-1b-SQL-Reasoning-GRPO-QLoRA
|
DeathReaper0965
| 2025-04-28T04:47:02Z | 0 | 0 |
transformers
|
[
"transformers",
"safetensors",
"generated_from_trainer",
"trl",
"grpo",
"text-generation",
"conversational",
"en",
"arxiv:2402.03300",
"base_model:google/gemma-3-1b-it",
"base_model:finetune:google/gemma-3-1b-it",
"license:mit",
"endpoints_compatible",
"region:us"
] |
text-generation
| 2025-04-28T04:14:12Z |
---
base_model: google/gemma-3-1b-it
library_name: transformers
model_name: Gemma-1b-SQL-Reasoning-GRPO-QLoRA
tags:
- generated_from_trainer
- trl
- grpo
licence: license
license: mit
language:
- en
pipeline_tag: text-generation
---
# Model Card for Gemma-1b-SQL-Reasoning-GRPO-QLoRA
This model is RL-tuned using GRPO to produce Reasoning based SQL Queries as an output.
## Quick start
```python
from transformers import pipeline
prompt = [
{
'role': 'system',
'content': """\
You are an expert SQL Query Writer.
Given relevant Schemas and the Question, you first understand the problem entirely and then reason about the best possible approach to come up with an answer.
Once, you are confident in your reasoning, you will then start generating the SQL Query as the answer that accurately solves the given question leveraging some or all schemas.
Remember that you should place all your reasoning between <reason> and </reason> tags.
Also, you should provide your solution between <answer> and </answer> tags.
An example generation is as follows:
<reason>
This is a sample reasoning that solves the question based on the schema.
</reason>
<answer>
SELECT
COLUMN
FROM TABLE_NAME
WHERE
CONDITION
</answer>"""
},
{
'role': 'user',
'content': """\
SCHEMAS:
---------------
CREATE TABLE Customers (
first_name VARCHAR,
last_name VARCHAR,
customer_id VARCHAR
)
CREATE TABLE Customer_Payments (
customer_id VARCHAR
)
---------------
QUESTION: "List first name and last name of customers that have more than 2 payments."
"""
}
]
generator = pipeline("text-generation", model="DeathReaper0965/Gemma-1b-SQL-Reasoning-GRPO-QLoRA", device="cuda")
output = generator(prompt, max_new_tokens=256, return_full_text=False)[0]
print(output["generated_text"])
###########OUTPUT###########
<reason>
The question asks to identify customers who have more than two payments. To achieve this, we need to filter the `Customers` table based on the `customer_id` and then select the `first_name` and `last_name` columns from the resulting filtered data. The `Customer_Payments` table is not relevant to this query, as it provides information about payments, not customer information. Therefore, we can directly query the `Customers` table. The logic is straightforward: select the `first_name` and `last_name` from the `Customers` table where `customer_id` appears more than once in the `Customer_Payments` table.
</reason>
<answer>
SELECT
first_name,
last_name
FROM
Customers
WHERE
customer_id IN (SELECT customer_id FROM Customer_Payments GROUP BY customer_id HAVING COUNT(*) > 2);
</answer>
```
## Training procedure
This model was trained with GRPO, a method introduced in [DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models](https://huggingface.co/papers/2402.03300).
### Framework versions
- TRL: 0.17.0
- Transformers: 4.51.3
- Pytorch: 2.6.0+cu124
- Datasets: 3.5.0
- Tokenizers: 0.21.1
## Citations
Cite GRPO as:
```bibtex
@article{zhihong2024deepseekmath,
title = {{DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models}},
author = {Zhihong Shao and Peiyi Wang and Qihao Zhu and Runxin Xu and Junxiao Song and Mingchuan Zhang and Y. K. Li and Y. Wu and Daya Guo},
year = 2024,
eprint = {arXiv:2402.03300},
}
```
Cite TRL as:
```bibtex
@misc{vonwerra2022trl,
title = {{TRL: Transformer Reinforcement Learning}},
author = {Leandro von Werra and Younes Belkada and Lewis Tunstall and Edward Beeching and Tristan Thrush and Nathan Lambert and Shengyi Huang and Kashif Rasul and Quentin Gallou{\'e}dec},
year = 2020,
journal = {GitHub repository},
publisher = {GitHub},
howpublished = {\url{https://github.com/huggingface/trl}}
}
```
> Designed and Developed with <span style="color: #e25555;">♥</span> by [Praneet](https://deathreaper0965.github.io/) | [LinkedIn](http://linkedin.com/in/deathreaper0965) | [GitHub](https://github.com/DeathReaper0965/)
|
rizkysulaeman/Gemma3-4B-multimodal-en-ft-v1-Q8_0-GGUF
|
rizkysulaeman
| 2025-04-28T04:46:51Z | 0 | 0 |
transformers
|
[
"transformers",
"gguf",
"text-generation-inference",
"unsloth",
"gemma3",
"llama-cpp",
"gguf-my-repo",
"en",
"base_model:CALISTA-INDUSTRY/Gemma3-4B-multimodal-en-ft-v1",
"base_model:quantized:CALISTA-INDUSTRY/Gemma3-4B-multimodal-en-ft-v1",
"license:apache-2.0",
"endpoints_compatible",
"region:us",
"conversational"
] | null | 2025-04-28T04:46:25Z |
---
base_model: CALISTA-INDUSTRY/Gemma3-4B-multimodal-en-ft-v1
language:
- en
license: apache-2.0
tags:
- text-generation-inference
- transformers
- unsloth
- gemma3
- llama-cpp
- gguf-my-repo
---
# rizkysulaeman/Gemma3-4B-multimodal-en-ft-v1-Q8_0-GGUF
This model was converted to GGUF format from [`CALISTA-INDUSTRY/Gemma3-4B-multimodal-en-ft-v1`](https://huggingface.co/CALISTA-INDUSTRY/Gemma3-4B-multimodal-en-ft-v1) using llama.cpp via the ggml.ai's [GGUF-my-repo](https://huggingface.co/spaces/ggml-org/gguf-my-repo) space.
Refer to the [original model card](https://huggingface.co/CALISTA-INDUSTRY/Gemma3-4B-multimodal-en-ft-v1) for more details on the model.
## Use with llama.cpp
Install llama.cpp through brew (works on Mac and Linux)
```bash
brew install llama.cpp
```
Invoke the llama.cpp server or the CLI.
### CLI:
```bash
llama-cli --hf-repo rizkysulaeman/Gemma3-4B-multimodal-en-ft-v1-Q8_0-GGUF --hf-file gemma3-4b-multimodal-en-ft-v1-q8_0.gguf -p "The meaning to life and the universe is"
```
### Server:
```bash
llama-server --hf-repo rizkysulaeman/Gemma3-4B-multimodal-en-ft-v1-Q8_0-GGUF --hf-file gemma3-4b-multimodal-en-ft-v1-q8_0.gguf -c 2048
```
Note: You can also use this checkpoint directly through the [usage steps](https://github.com/ggerganov/llama.cpp?tab=readme-ov-file#usage) listed in the Llama.cpp repo as well.
Step 1: Clone llama.cpp from GitHub.
```
git clone https://github.com/ggerganov/llama.cpp
```
Step 2: Move into the llama.cpp folder and build it with `LLAMA_CURL=1` flag along with other hardware-specific flags (for ex: LLAMA_CUDA=1 for Nvidia GPUs on Linux).
```
cd llama.cpp && LLAMA_CURL=1 make
```
Step 3: Run inference through the main binary.
```
./llama-cli --hf-repo rizkysulaeman/Gemma3-4B-multimodal-en-ft-v1-Q8_0-GGUF --hf-file gemma3-4b-multimodal-en-ft-v1-q8_0.gguf -p "The meaning to life and the universe is"
```
or
```
./llama-server --hf-repo rizkysulaeman/Gemma3-4B-multimodal-en-ft-v1-Q8_0-GGUF --hf-file gemma3-4b-multimodal-en-ft-v1-q8_0.gguf -c 2048
```
|
mradermacher/S1-1k-32B-GGUF
|
mradermacher
| 2025-04-28T04:43:27Z | 0 | 0 |
transformers
|
[
"transformers",
"gguf",
"en",
"base_model:InfiniAILab/S1-1k-32B",
"base_model:quantized:InfiniAILab/S1-1k-32B",
"endpoints_compatible",
"region:us",
"conversational"
] | null | 2025-04-27T22:39:35Z |
---
base_model: InfiniAILab/S1-1k-32B
language:
- en
library_name: transformers
quantized_by: mradermacher
---
## About
<!-- ### quantize_version: 2 -->
<!-- ### output_tensor_quantised: 1 -->
<!-- ### convert_type: hf -->
<!-- ### vocab_type: -->
<!-- ### tags: -->
static quants of https://huggingface.co/InfiniAILab/S1-1k-32B
<!-- provided-files -->
weighted/imatrix quants are available at https://huggingface.co/mradermacher/S1-1k-32B-i1-GGUF
## Usage
If you are unsure how to use GGUF files, refer to one of [TheBloke's
READMEs](https://huggingface.co/TheBloke/KafkaLM-70B-German-V0.1-GGUF) for
more details, including on how to concatenate multi-part files.
## Provided Quants
(sorted by size, not necessarily quality. IQ-quants are often preferable over similar sized non-IQ quants)
| Link | Type | Size/GB | Notes |
|:-----|:-----|--------:|:------|
| [GGUF](https://huggingface.co/mradermacher/S1-1k-32B-GGUF/resolve/main/S1-1k-32B.Q2_K.gguf) | Q2_K | 12.4 | |
| [GGUF](https://huggingface.co/mradermacher/S1-1k-32B-GGUF/resolve/main/S1-1k-32B.Q3_K_S.gguf) | Q3_K_S | 14.5 | |
| [GGUF](https://huggingface.co/mradermacher/S1-1k-32B-GGUF/resolve/main/S1-1k-32B.Q3_K_M.gguf) | Q3_K_M | 16.0 | lower quality |
| [GGUF](https://huggingface.co/mradermacher/S1-1k-32B-GGUF/resolve/main/S1-1k-32B.Q3_K_L.gguf) | Q3_K_L | 17.3 | |
| [GGUF](https://huggingface.co/mradermacher/S1-1k-32B-GGUF/resolve/main/S1-1k-32B.IQ4_XS.gguf) | IQ4_XS | 18.0 | |
| [GGUF](https://huggingface.co/mradermacher/S1-1k-32B-GGUF/resolve/main/S1-1k-32B.Q4_K_S.gguf) | Q4_K_S | 18.9 | fast, recommended |
| [GGUF](https://huggingface.co/mradermacher/S1-1k-32B-GGUF/resolve/main/S1-1k-32B.Q4_K_M.gguf) | Q4_K_M | 20.0 | fast, recommended |
| [GGUF](https://huggingface.co/mradermacher/S1-1k-32B-GGUF/resolve/main/S1-1k-32B.Q5_K_S.gguf) | Q5_K_S | 22.7 | |
| [GGUF](https://huggingface.co/mradermacher/S1-1k-32B-GGUF/resolve/main/S1-1k-32B.Q5_K_M.gguf) | Q5_K_M | 23.4 | |
| [GGUF](https://huggingface.co/mradermacher/S1-1k-32B-GGUF/resolve/main/S1-1k-32B.Q6_K.gguf) | Q6_K | 27.0 | very good quality |
| [GGUF](https://huggingface.co/mradermacher/S1-1k-32B-GGUF/resolve/main/S1-1k-32B.Q8_0.gguf) | Q8_0 | 34.9 | fast, best quality |
Here is a handy graph by ikawrakow comparing some lower-quality quant
types (lower is better):
And here are Artefact2's thoughts on the matter:
https://gist.github.com/Artefact2/b5f810600771265fc1e39442288e8ec9
## FAQ / Model Request
See https://huggingface.co/mradermacher/model_requests for some answers to
questions you might have and/or if you want some other model quantized.
## Thanks
I thank my company, [nethype GmbH](https://www.nethype.de/), for letting
me use its servers and providing upgrades to my workstation to enable
this work in my free time.
<!-- end -->
|
mradermacher/Qwen-Qwen2.5-1.5B-llamafied-GGUF
|
mradermacher
| 2025-04-28T04:43:13Z | 28 | 0 |
transformers
|
[
"transformers",
"gguf",
"zho",
"eng",
"fra",
"spa",
"por",
"deu",
"ita",
"rus",
"jpn",
"kor",
"vie",
"tha",
"ara",
"base_model:llamafy/Qwen-Qwen2.5-1.5B-llamafied",
"base_model:quantized:llamafy/Qwen-Qwen2.5-1.5B-llamafied",
"endpoints_compatible",
"region:us",
"conversational"
] | null | 2024-11-17T05:27:00Z |
---
base_model: llamafy/Qwen-Qwen2.5-1.5B-llamafied
language:
- zho
- eng
- fra
- spa
- por
- deu
- ita
- rus
- jpn
- kor
- vie
- tha
- ara
library_name: transformers
quantized_by: mradermacher
---
## About
<!-- ### quantize_version: 2 -->
<!-- ### output_tensor_quantised: 1 -->
<!-- ### convert_type: hf -->
<!-- ### vocab_type: -->
<!-- ### tags: -->
static quants of https://huggingface.co/llamafy/Qwen-Qwen2.5-1.5B-llamafied
<!-- provided-files -->
weighted/imatrix quants seem not to be available (by me) at this time. If they do not show up a week or so after the static ones, I have probably not planned for them. Feel free to request them by opening a Community Discussion.
## Usage
If you are unsure how to use GGUF files, refer to one of [TheBloke's
READMEs](https://huggingface.co/TheBloke/KafkaLM-70B-German-V0.1-GGUF) for
more details, including on how to concatenate multi-part files.
## Provided Quants
(sorted by size, not necessarily quality. IQ-quants are often preferable over similar sized non-IQ quants)
| Link | Type | Size/GB | Notes |
|:-----|:-----|--------:|:------|
| [GGUF](https://huggingface.co/mradermacher/Qwen-Qwen2.5-1.5B-llamafied-GGUF/resolve/main/Qwen-Qwen2.5-1.5B-llamafied.Q2_K.gguf) | Q2_K | 0.8 | |
| [GGUF](https://huggingface.co/mradermacher/Qwen-Qwen2.5-1.5B-llamafied-GGUF/resolve/main/Qwen-Qwen2.5-1.5B-llamafied.Q3_K_S.gguf) | Q3_K_S | 0.9 | |
| [GGUF](https://huggingface.co/mradermacher/Qwen-Qwen2.5-1.5B-llamafied-GGUF/resolve/main/Qwen-Qwen2.5-1.5B-llamafied.Q3_K_M.gguf) | Q3_K_M | 0.9 | lower quality |
| [GGUF](https://huggingface.co/mradermacher/Qwen-Qwen2.5-1.5B-llamafied-GGUF/resolve/main/Qwen-Qwen2.5-1.5B-llamafied.Q3_K_L.gguf) | Q3_K_L | 1.0 | |
| [GGUF](https://huggingface.co/mradermacher/Qwen-Qwen2.5-1.5B-llamafied-GGUF/resolve/main/Qwen-Qwen2.5-1.5B-llamafied.IQ4_XS.gguf) | IQ4_XS | 1.0 | |
| [GGUF](https://huggingface.co/mradermacher/Qwen-Qwen2.5-1.5B-llamafied-GGUF/resolve/main/Qwen-Qwen2.5-1.5B-llamafied.Q4_0_4_4.gguf) | Q4_0_4_4 | 1.0 | fast on arm, low quality |
| [GGUF](https://huggingface.co/mradermacher/Qwen-Qwen2.5-1.5B-llamafied-GGUF/resolve/main/Qwen-Qwen2.5-1.5B-llamafied.Q4_K_S.gguf) | Q4_K_S | 1.0 | fast, recommended |
| [GGUF](https://huggingface.co/mradermacher/Qwen-Qwen2.5-1.5B-llamafied-GGUF/resolve/main/Qwen-Qwen2.5-1.5B-llamafied.Q4_K_M.gguf) | Q4_K_M | 1.1 | fast, recommended |
| [GGUF](https://huggingface.co/mradermacher/Qwen-Qwen2.5-1.5B-llamafied-GGUF/resolve/main/Qwen-Qwen2.5-1.5B-llamafied.Q5_K_S.gguf) | Q5_K_S | 1.2 | |
| [GGUF](https://huggingface.co/mradermacher/Qwen-Qwen2.5-1.5B-llamafied-GGUF/resolve/main/Qwen-Qwen2.5-1.5B-llamafied.Q5_K_M.gguf) | Q5_K_M | 1.2 | |
| [GGUF](https://huggingface.co/mradermacher/Qwen-Qwen2.5-1.5B-llamafied-GGUF/resolve/main/Qwen-Qwen2.5-1.5B-llamafied.Q6_K.gguf) | Q6_K | 1.4 | very good quality |
| [GGUF](https://huggingface.co/mradermacher/Qwen-Qwen2.5-1.5B-llamafied-GGUF/resolve/main/Qwen-Qwen2.5-1.5B-llamafied.Q8_0.gguf) | Q8_0 | 1.7 | fast, best quality |
| [GGUF](https://huggingface.co/mradermacher/Qwen-Qwen2.5-1.5B-llamafied-GGUF/resolve/main/Qwen-Qwen2.5-1.5B-llamafied.f16.gguf) | f16 | 3.2 | 16 bpw, overkill |
Here is a handy graph by ikawrakow comparing some lower-quality quant
types (lower is better):
And here are Artefact2's thoughts on the matter:
https://gist.github.com/Artefact2/b5f810600771265fc1e39442288e8ec9
## FAQ / Model Request
See https://huggingface.co/mradermacher/model_requests for some answers to
questions you might have and/or if you want some other model quantized.
## Thanks
I thank my company, [nethype GmbH](https://www.nethype.de/), for letting
me use its servers and providing upgrades to my workstation to enable
this work in my free time.
<!-- end -->
|
18-Jobz-Hunting-Sajal-Malik-Go-Viral/Full.Clip.Jobz.Hunting.Sajal.Malik.Viral.Video.Leaks.official
|
18-Jobz-Hunting-Sajal-Malik-Go-Viral
| 2025-04-28T04:42:38Z | 0 | 0 | null |
[
"region:us"
] | null | 2025-04-28T04:37:54Z |
<animated-image data-catalyst=""><a href=" https://tinyurl.com/5n7shfr3?dfhgKasbonStudiosdfg" rel="nofollow" data-target="animated-image.originalLink"><img src="https://static.wixstatic.com/media/b249f9_adac8f70fb3f45b88691696c77de18f3~mv2.gif" alt="Foo" data-canonical-src="https://static.wixstatic.com/media/b249f9_adac8f70fb3f45b88691696c77de18f3~mv2.gif" style="max-width: 100%; display: inline-block;" data-target="animated-image.originalImage"></a>
Actor jobz hunting sajal malik Original V𝚒deo V𝚒deo took the internet by storm and amazed viewers on various social media platforms. Actor jobz hunting sajal malik, a young and talented digital creator, recently became famous thanks to this interesting V𝚒deo.
L𝚎aked V𝚒deo Actor jobz hunting sajal malik V𝚒ral V𝚒deo Original V𝚒deo L𝚒nk On Social Media Telegram X Trending Tiktok (18+)
L𝚎aked V𝚒deo Actor jobz hunting sajal malik V𝚒ral V𝚒deo Original V𝚒deo L𝚒nk On Social Media X Trending Tiktok (18+)
L𝚎aked V𝚒deo Actor jobz hunting sajal malik Original V𝚒deo V𝚒ral V𝚒deo L𝚎aked on X Twitter
|
aicouture/blue_style
|
aicouture
| 2025-04-28T04:40:53Z | 0 | 0 |
diffusers
|
[
"diffusers",
"flux",
"lora",
"replicate",
"text-to-image",
"en",
"base_model:black-forest-labs/FLUX.1-dev",
"base_model:adapter:black-forest-labs/FLUX.1-dev",
"license:other",
"region:us"
] |
text-to-image
| 2025-04-28T04:24:45Z |
---
license: other
license_name: flux-1-dev-non-commercial-license
license_link: https://huggingface.co/black-forest-labs/FLUX.1-dev/blob/main/LICENSE.md
language:
- en
tags:
- flux
- diffusers
- lora
- replicate
base_model: "black-forest-labs/FLUX.1-dev"
pipeline_tag: text-to-image
# widget:
# - text: >-
# prompt
# output:
# url: https://...
instance_prompt: BLUESTYLE
---
# Blue_Style
<Gallery />
## About this LoRA
This is a [LoRA](https://replicate.com/docs/guides/working-with-loras) for the FLUX.1-dev text-to-image model. It can be used with diffusers or ComfyUI.
It was trained on [Replicate](https://replicate.com/) using AI toolkit: https://replicate.com/ostris/flux-dev-lora-trainer/train
## Trigger words
You should use `BLUESTYLE` to trigger the image generation.
## Run this LoRA with an API using Replicate
```py
import replicate
input = {
"prompt": "BLUESTYLE",
"lora_weights": "https://huggingface.co/aicouture/blue_style/resolve/main/lora.safetensors"
}
output = replicate.run(
"black-forest-labs/flux-dev-lora",
input=input
)
for index, item in enumerate(output):
with open(f"output_{index}.webp", "wb") as file:
file.write(item.read())
```
## Use it with the [🧨 diffusers library](https://github.com/huggingface/diffusers)
```py
from diffusers import AutoPipelineForText2Image
import torch
pipeline = AutoPipelineForText2Image.from_pretrained('black-forest-labs/FLUX.1-dev', torch_dtype=torch.float16).to('cuda')
pipeline.load_lora_weights('aicouture/blue_style', weight_name='lora.safetensors')
image = pipeline('BLUESTYLE').images[0]
```
For more details, including weighting, merging and fusing LoRAs, check the [documentation on loading LoRAs in diffusers](https://huggingface.co/docs/diffusers/main/en/using-diffusers/loading_adapters)
## Training details
- Steps: 1000
- Learning rate: 0.0004
- LoRA rank: 16
## Contribute your own examples
You can use the [community tab](https://huggingface.co/aicouture/blue_style/discussions) to add images that show off what you’ve made with this LoRA.
|
TOMFORD79/S1
|
TOMFORD79
| 2025-04-28T04:34:41Z | 0 | 0 | null |
[
"safetensors",
"any-to-any",
"omega",
"omegalabs",
"bittensor",
"agi",
"license:mit",
"region:us"
] |
any-to-any
| 2025-04-28T04:02:02Z |
---
license: mit
tags:
- any-to-any
- omega
- omegalabs
- bittensor
- agi
---
This is an Any-to-Any model checkpoint for the OMEGA Labs x Bittensor Any-to-Any subnet.
Check out the [git repo](https://github.com/omegalabsinc/omegalabs-anytoany-bittensor) and find OMEGA on X: [@omegalabsai](https://x.com/omegalabsai).
|
archieto0412/whisper-atc-finetuned
|
archieto0412
| 2025-04-28T04:33:32Z | 20 | 0 |
transformers
|
[
"transformers",
"tensorboard",
"safetensors",
"whisper",
"automatic-speech-recognition",
"generated_from_trainer",
"en",
"dataset:archieto0412/atc-recordings",
"base_model:openai/whisper-small",
"base_model:finetune:openai/whisper-small",
"license:apache-2.0",
"model-index",
"endpoints_compatible",
"region:us"
] |
automatic-speech-recognition
| 2025-04-13T17:36:05Z |
---
library_name: transformers
language:
- en
license: apache-2.0
base_model: openai/whisper-small
tags:
- generated_from_trainer
datasets:
- archieto0412/atc-recordings
metrics:
- wer
model-index:
- name: Fine-tuned Whisper for AM
results:
- task:
name: Automatic Speech Recognition
type: automatic-speech-recognition
dataset:
name: ATC voice recordings
type: archieto0412/atc-recordings
config: default
split: validation
args: default
metrics:
- name: Wer
type: wer
value: 0.18688230008984727
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# Fine-tuned Whisper for AM
This model is a fine-tuned version of [openai/whisper-small](https://huggingface.co/openai/whisper-small) on the ATC voice recordings dataset.
It achieves the following results on the evaluation set:
- Loss: 0.5964
- Wer: 0.1869
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 5e-06
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 20
- training_steps: 400
- mixed_precision_training: Native AMP
### Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|:-------------:|:-------:|:----:|:---------------:|:------:|
| 1.3484 | 2.2727 | 50 | 1.0828 | 0.1986 |
| 0.7309 | 4.5455 | 100 | 0.9440 | 0.1752 |
| 0.5176 | 6.8182 | 150 | 0.9446 | 0.1869 |
| 0.2552 | 9.0909 | 200 | 0.5440 | 0.1923 |
| 0.0315 | 11.3636 | 250 | 0.5883 | 0.1959 |
| 0.0069 | 13.6364 | 300 | 0.5849 | 0.1842 |
| 0.0041 | 15.9091 | 350 | 0.5954 | 0.1851 |
| 0.0033 | 18.1818 | 400 | 0.5964 | 0.1869 |
### Framework versions
- Transformers 4.52.0.dev0
- Pytorch 2.6.0+cu124
- Datasets 3.5.0
- Tokenizers 0.21.1
|
b5Vza9kApSQ/hjdagaf
|
b5Vza9kApSQ
| 2025-04-28T04:33:17Z | 0 | 0 | null |
[
"license:apache-2.0",
"region:us"
] | null | 2025-04-28T04:33:16Z |
---
license: apache-2.0
---
|
aleegis/ec3851b9-4056-4247-98af-b83d2a5be1c8
|
aleegis
| 2025-04-28T04:33:04Z | 0 | 0 |
peft
|
[
"peft",
"safetensors",
"qwen2",
"axolotl",
"generated_from_trainer",
"base_model:Qwen/Qwen2.5-1.5B-Instruct",
"base_model:adapter:Qwen/Qwen2.5-1.5B-Instruct",
"license:apache-2.0",
"region:us"
] | null | 2025-04-28T03:58:27Z |
---
library_name: peft
license: apache-2.0
base_model: Qwen/Qwen2.5-1.5B-Instruct
tags:
- axolotl
- generated_from_trainer
model-index:
- name: ec3851b9-4056-4247-98af-b83d2a5be1c8
results: []
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
[<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl)
<details><summary>See axolotl config</summary>
axolotl version: `0.4.1`
```yaml
adapter: lora
base_model: Qwen/Qwen2.5-1.5B-Instruct
bf16: auto
chat_template: llama3
dataloader_num_workers: 12
dataset_prepared_path: null
datasets:
- data_files:
- f2392decb627cf18_train_data.json
ds_type: json
format: custom
path: /workspace/input_data/f2392decb627cf18_train_data.json
type:
field_input: statements
field_instruction: quiz
field_output: solution_text
format: '{instruction} {input}'
no_input_format: '{instruction}'
system_format: '{system}'
system_prompt: ''
debug: null
deepspeed: null
early_stopping_patience: null
eval_max_new_tokens: 128
eval_steps: null
eval_table_size: null
evals_per_epoch: null
flash_attention: true
fp16: null
fsdp: null
fsdp_config: null
gradient_accumulation_steps: 8
gradient_checkpointing: false
group_by_length: false
hub_model_id: aleegis/ec3851b9-4056-4247-98af-b83d2a5be1c8
hub_repo: null
hub_strategy: checkpoint
hub_token: null
learning_rate: 0.0001
load_in_4bit: false
load_in_8bit: false
local_rank: null
logging_steps: null
lora_alpha: 32
lora_dropout: 0.15
lora_fan_in_fan_out: null
lora_model_dir: null
lora_r: 32
lora_target_linear: true
loraplus_lr_embedding: 1.0e-06
loraplus_lr_ratio: 16
lr_scheduler: cosine
max_grad_norm: 1
max_steps: 1500
micro_batch_size: 2
mlflow_experiment_name: /tmp/f2392decb627cf18_train_data.json
model_type: AutoModelForCausalLM
num_epochs: 200
optimizer: adamw_torch_fused
output_dir: miner_id_24
pad_to_sequence_len: true
resume_from_checkpoint: null
s2_attention: null
sample_packing: false
save_steps: null
save_total_limit: 10
saves_per_epoch: 0
sequence_len: 1024
strict: false
tf32: true
tokenizer_type: AutoTokenizer
train_on_inputs: false
trust_remote_code: true
val_set_size: 0.0
wandb_entity: null
wandb_mode: online
wandb_name: a54f4409-dd56-46d7-8e17-1d233ee1e00a
wandb_project: Gradients-On-Demand
wandb_run: your_name
wandb_runid: a54f4409-dd56-46d7-8e17-1d233ee1e00a
warmup_steps: 100
weight_decay: 0
xformers_attention: null
```
</details><br>
# ec3851b9-4056-4247-98af-b83d2a5be1c8
This model is a fine-tuned version of [Qwen/Qwen2.5-1.5B-Instruct](https://huggingface.co/Qwen/Qwen2.5-1.5B-Instruct) on the None dataset.
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 0.0001
- train_batch_size: 2
- eval_batch_size: 2
- seed: 42
- gradient_accumulation_steps: 8
- total_train_batch_size: 16
- optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: cosine
- lr_scheduler_warmup_steps: 100
- training_steps: 1500
### Training results
### Framework versions
- PEFT 0.13.2
- Transformers 4.46.0
- Pytorch 2.5.0+cu124
- Datasets 3.0.1
- Tokenizers 0.20.1
|
nqdhocai/LogicLlama-3.2-1B-NoDes-v0
|
nqdhocai
| 2025-04-28T04:29:01Z | 0 | 0 |
transformers
|
[
"transformers",
"safetensors",
"llama",
"text-generation",
"text-generation-inference",
"unsloth",
"trl",
"sft",
"conversational",
"en",
"base_model:unsloth/Llama-3.2-1B-Instruct",
"base_model:finetune:unsloth/Llama-3.2-1B-Instruct",
"license:apache-2.0",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] |
text-generation
| 2025-04-28T04:27:45Z |
---
base_model: unsloth/Llama-3.2-1B-Instruct
tags:
- text-generation-inference
- transformers
- unsloth
- llama
- trl
- sft
license: apache-2.0
language:
- en
---
# Uploaded model
- **Developed by:** nqdhocai
- **License:** apache-2.0
- **Finetuned from model :** unsloth/Llama-3.2-1B-Instruct
This llama model was trained 2x faster with [Unsloth](https://github.com/unslothai/unsloth) and Huggingface's TRL library.
[<img src="https://raw.githubusercontent.com/unslothai/unsloth/main/images/unsloth%20made%20with%20love.png" width="200"/>](https://github.com/unslothai/unsloth)
|
Williams10312/medical-question-model
|
Williams10312
| 2025-04-28T04:27:30Z | 0 | 0 |
transformers
|
[
"transformers",
"safetensors",
"bert",
"text-classification",
"arxiv:1910.09700",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] |
text-classification
| 2025-04-28T04:27:15Z |
---
library_name: transformers
tags: []
---
# Model Card for Model ID
<!-- Provide a quick summary of what the model is/does. -->
## Model Details
### Model Description
<!-- Provide a longer summary of what this model is. -->
This is the model card of a 🤗 transformers model that has been pushed on the Hub. This model card has been automatically generated.
- **Developed by:** [More Information Needed]
- **Funded by [optional]:** [More Information Needed]
- **Shared by [optional]:** [More Information Needed]
- **Model type:** [More Information Needed]
- **Language(s) (NLP):** [More Information Needed]
- **License:** [More Information Needed]
- **Finetuned from model [optional]:** [More Information Needed]
### Model Sources [optional]
<!-- Provide the basic links for the model. -->
- **Repository:** [More Information Needed]
- **Paper [optional]:** [More Information Needed]
- **Demo [optional]:** [More Information Needed]
## Uses
<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
### Direct Use
<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
[More Information Needed]
### Downstream Use [optional]
<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
[More Information Needed]
### Out-of-Scope Use
<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
[More Information Needed]
## Bias, Risks, and Limitations
<!-- This section is meant to convey both technical and sociotechnical limitations. -->
[More Information Needed]
### Recommendations
<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
## How to Get Started with the Model
Use the code below to get started with the model.
[More Information Needed]
## Training Details
### Training Data
<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
[More Information Needed]
### Training Procedure
<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
#### Preprocessing [optional]
[More Information Needed]
#### Training Hyperparameters
- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
#### Speeds, Sizes, Times [optional]
<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
[More Information Needed]
## Evaluation
<!-- This section describes the evaluation protocols and provides the results. -->
### Testing Data, Factors & Metrics
#### Testing Data
<!-- This should link to a Dataset Card if possible. -->
[More Information Needed]
#### Factors
<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
[More Information Needed]
#### Metrics
<!-- These are the evaluation metrics being used, ideally with a description of why. -->
[More Information Needed]
### Results
[More Information Needed]
#### Summary
## Model Examination [optional]
<!-- Relevant interpretability work for the model goes here -->
[More Information Needed]
## Environmental Impact
<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
- **Hardware Type:** [More Information Needed]
- **Hours used:** [More Information Needed]
- **Cloud Provider:** [More Information Needed]
- **Compute Region:** [More Information Needed]
- **Carbon Emitted:** [More Information Needed]
## Technical Specifications [optional]
### Model Architecture and Objective
[More Information Needed]
### Compute Infrastructure
[More Information Needed]
#### Hardware
[More Information Needed]
#### Software
[More Information Needed]
## Citation [optional]
<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
**BibTeX:**
[More Information Needed]
**APA:**
[More Information Needed]
## Glossary [optional]
<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
[More Information Needed]
## More Information [optional]
[More Information Needed]
## Model Card Authors [optional]
[More Information Needed]
## Model Card Contact
[More Information Needed]
|
EasierAI/Qwen-2.5-3B
|
EasierAI
| 2025-04-28T04:20:38Z | 59 | 0 | null |
[
"gguf",
"text-generation",
"zho",
"eng",
"fra",
"spa",
"por",
"deu",
"ita",
"rus",
"jpn",
"kor",
"vie",
"tha",
"ara",
"base_model:Qwen/Qwen2.5-3B",
"base_model:quantized:Qwen/Qwen2.5-3B",
"endpoints_compatible",
"region:us",
"conversational"
] |
text-generation
| 2025-02-13T17:02:36Z |
---
base_model: Qwen/Qwen2.5-3B
pipeline_tag: text-generation
quantized_by: bartowski
language:
- zho
- eng
- fra
- spa
- por
- deu
- ita
- rus
- jpn
- kor
- vie
- tha
- ara
---
## Llamacpp imatrix Quantizations of Qwen2.5-3B
Using <a href="https://github.com/ggerganov/llama.cpp/">llama.cpp</a> release <a href="https://github.com/ggerganov/llama.cpp/releases/tag/b3751">b3751</a> for quantization.
Original model: https://huggingface.co/Qwen/Qwen2.5-3B
All quants made using imatrix option with dataset from [here](https://gist.github.com/bartowski1182/eb213dccb3571f863da82e99418f81e8)
Run them in [LM Studio](https://lmstudio.ai/)
## Prompt format
```
<|im_start|>system
{system_prompt}<|im_end|>
<|im_start|>user
{prompt}<|im_end|>
<|im_start|>assistant
```
## Download a file (not the whole branch) from below:
| Filename | Quant type | File Size | Split | Description |
| -------- | ---------- | --------- | ----- | ----------- |
| [Qwen2.5-3B-f32.gguf](https://huggingface.co/bartowski/Qwen2.5-3B-GGUF/blob/main/Qwen2.5-3B-f32.gguf) | f32 | 12.35GB | false | Full F32 weights. |
| [Qwen2.5-3B-Q8_0.gguf](https://huggingface.co/bartowski/Qwen2.5-3B-GGUF/blob/main/Qwen2.5-3B-Q8_0.gguf) | Q8_0 | 3.29GB | false | Extremely high quality, generally unneeded but max available quant. |
| [Qwen2.5-3B-Q6_K_L.gguf](https://huggingface.co/bartowski/Qwen2.5-3B-GGUF/blob/main/Qwen2.5-3B-Q6_K_L.gguf) | Q6_K_L | 2.61GB | false | Uses Q8_0 for embed and output weights. Very high quality, near perfect, *recommended*. |
| [Qwen2.5-3B-Q6_K.gguf](https://huggingface.co/bartowski/Qwen2.5-3B-GGUF/blob/main/Qwen2.5-3B-Q6_K.gguf) | Q6_K | 2.54GB | false | Very high quality, near perfect, *recommended*. |
| [Qwen2.5-3B-Q5_K_L.gguf](https://huggingface.co/bartowski/Qwen2.5-3B-GGUF/blob/main/Qwen2.5-3B-Q5_K_L.gguf) | Q5_K_L | 2.30GB | false | Uses Q8_0 for embed and output weights. High quality, *recommended*. |
| [Qwen2.5-3B-Q5_K_M.gguf](https://huggingface.co/bartowski/Qwen2.5-3B-GGUF/blob/main/Qwen2.5-3B-Q5_K_M.gguf) | Q5_K_M | 2.22GB | false | High quality, *recommended*. |
| [Qwen2.5-3B-Q5_K_S.gguf](https://huggingface.co/bartowski/Qwen2.5-3B-GGUF/blob/main/Qwen2.5-3B-Q5_K_S.gguf) | Q5_K_S | 2.17GB | false | High quality, *recommended*. |
| [Qwen2.5-3B-Q4_K_L.gguf](https://huggingface.co/bartowski/Qwen2.5-3B-GGUF/blob/main/Qwen2.5-3B-Q4_K_L.gguf) | Q4_K_L | 2.01GB | false | Uses Q8_0 for embed and output weights. Good quality, *recommended*. |
| [Qwen2.5-3B-Q4_K_M.gguf](https://huggingface.co/bartowski/Qwen2.5-3B-GGUF/blob/main/Qwen2.5-3B-Q4_K_M.gguf) | Q4_K_M | 1.93GB | false | Good quality, default size for must use cases, *recommended*. |
| [Qwen2.5-3B-Q4_K_S.gguf](https://huggingface.co/bartowski/Qwen2.5-3B-GGUF/blob/main/Qwen2.5-3B-Q4_K_S.gguf) | Q4_K_S | 1.83GB | false | Slightly lower quality with more space savings, *recommended*. |
| [Qwen2.5-3B-Q4_0.gguf](https://huggingface.co/bartowski/Qwen2.5-3B-GGUF/blob/main/Qwen2.5-3B-Q4_0.gguf) | Q4_0 | 1.83GB | false | Legacy format, generally not worth using over similarly sized formats |
| [Qwen2.5-3B-Q4_0_8_8.gguf](https://huggingface.co/bartowski/Qwen2.5-3B-GGUF/blob/main/Qwen2.5-3B-Q4_0_8_8.gguf) | Q4_0_8_8 | 1.82GB | false | Optimized for ARM inference. Requires 'sve' support (see link below). |
| [Qwen2.5-3B-Q4_0_4_8.gguf](https://huggingface.co/bartowski/Qwen2.5-3B-GGUF/blob/main/Qwen2.5-3B-Q4_0_4_8.gguf) | Q4_0_4_8 | 1.82GB | false | Optimized for ARM inference. Requires 'i8mm' support (see link below). |
| [Qwen2.5-3B-Q4_0_4_4.gguf](https://huggingface.co/bartowski/Qwen2.5-3B-GGUF/blob/main/Qwen2.5-3B-Q4_0_4_4.gguf) | Q4_0_4_4 | 1.82GB | false | Optimized for ARM inference. Should work well on all ARM chips, pick this if you're unsure. |
| [Qwen2.5-3B-Q3_K_XL.gguf](https://huggingface.co/bartowski/Qwen2.5-3B-GGUF/blob/main/Qwen2.5-3B-Q3_K_XL.gguf) | Q3_K_XL | 1.78GB | false | Uses Q8_0 for embed and output weights. Lower quality but usable, good for low RAM availability. |
| [Qwen2.5-3B-IQ4_XS.gguf](https://huggingface.co/bartowski/Qwen2.5-3B-GGUF/blob/main/Qwen2.5-3B-IQ4_XS.gguf) | IQ4_XS | 1.74GB | false | Decent quality, smaller than Q4_K_S with similar performance, *recommended*. |
| [Qwen2.5-3B-Q3_K_L.gguf](https://huggingface.co/bartowski/Qwen2.5-3B-GGUF/blob/main/Qwen2.5-3B-Q3_K_L.gguf) | Q3_K_L | 1.71GB | false | Lower quality but usable, good for low RAM availability. |
| [Qwen2.5-3B-IQ3_M.gguf](https://huggingface.co/bartowski/Qwen2.5-3B-GGUF/blob/main/Qwen2.5-3B-IQ3_M.gguf) | IQ3_M | 1.49GB | false | Medium-low quality, new method with decent performance comparable to Q3_K_M. |
## Embed/output weights
Some of these quants (Q3_K_XL, Q4_K_L etc) are the standard quantization method with the embeddings and output weights quantized to Q8_0 instead of what they would normally default to.
Some say that this improves the quality, others don't notice any difference. If you use these models PLEASE COMMENT with your findings. I would like feedback that these are actually used and useful so I don't keep uploading quants no one is using.
Thanks!
## Downloading using huggingface-cli
First, make sure you have hugginface-cli installed:
```
pip install -U "huggingface_hub[cli]"
```
Then, you can target the specific file you want:
```
huggingface-cli download bartowski/Qwen2.5-3B-GGUF --include "Qwen2.5-3B-Q4_K_M.gguf" --local-dir ./
```
If the model is bigger than 50GB, it will have been split into multiple files. In order to download them all to a local folder, run:
```
huggingface-cli download bartowski/Qwen2.5-3B-GGUF --include "Qwen2.5-3B-Q8_0/*" --local-dir ./
```
You can either specify a new local-dir (Qwen2.5-3B-Q8_0) or download them all in place (./)
## Q4_0_X_X
These are *NOT* for Metal (Apple) offloading, only ARM chips.
If you're using an ARM chip, the Q4_0_X_X quants will have a substantial speedup. Check out Q4_0_4_4 speed comparisons [on the original pull request](https://github.com/ggerganov/llama.cpp/pull/5780#pullrequestreview-21657544660)
To check which one would work best for your ARM chip, you can check [AArch64 SoC features](https://gpages.juszkiewicz.com.pl/arm-socs-table/arm-socs.html) (thanks EloyOn!).
## Which file should I choose?
A great write up with charts showing various performances is provided by Artefact2 [here](https://gist.github.com/Artefact2/b5f810600771265fc1e39442288e8ec9)
The first thing to figure out is how big a model you can run. To do this, you'll need to figure out how much RAM and/or VRAM you have.
If you want your model running as FAST as possible, you'll want to fit the whole thing on your GPU's VRAM. Aim for a quant with a file size 1-2GB smaller than your GPU's total VRAM.
If you want the absolute maximum quality, add both your system RAM and your GPU's VRAM together, then similarly grab a quant with a file size 1-2GB Smaller than that total.
Next, you'll need to decide if you want to use an 'I-quant' or a 'K-quant'.
If you don't want to think too much, grab one of the K-quants. These are in format 'QX_K_X', like Q5_K_M.
If you want to get more into the weeds, you can check out this extremely useful feature chart:
[llama.cpp feature matrix](https://github.com/ggerganov/llama.cpp/wiki/Feature-matrix)
But basically, if you're aiming for below Q4, and you're running cuBLAS (Nvidia) or rocBLAS (AMD), you should look towards the I-quants. These are in format IQX_X, like IQ3_M. These are newer and offer better performance for their size.
These I-quants can also be used on CPU and Apple Metal, but will be slower than their K-quant equivalent, so speed vs performance is a tradeoff you'll have to decide.
The I-quants are *not* compatible with Vulcan, which is also AMD, so if you have an AMD card double check if you're using the rocBLAS build or the Vulcan build. At the time of writing this, LM Studio has a preview with ROCm support, and other inference engines have specific builds for ROCm.
## Credits
Thank you kalomaze and Dampf for assistance in creating the imatrix calibration dataset
Thank you ZeroWw for the inspiration to experiment with embed/output
Want to support my work? Visit my ko-fi page here: https://ko-fi.com/bartowski
|
suriacaa/Qwen2.5-0.5B-Instruct-Gensyn-Swarm-quiet_shaggy_skunk
|
suriacaa
| 2025-04-28T04:19:49Z | 0 | 0 |
transformers
|
[
"transformers",
"safetensors",
"qwen2",
"text-generation",
"generated_from_trainer",
"rl-swarm",
"grpo",
"gensyn",
"I am quiet shaggy skunk",
"trl",
"conversational",
"arxiv:2402.03300",
"base_model:Gensyn/Qwen2.5-0.5B-Instruct",
"base_model:finetune:Gensyn/Qwen2.5-0.5B-Instruct",
"autotrain_compatible",
"text-generation-inference",
"endpoints_compatible",
"region:us"
] |
text-generation
| 2025-04-28T03:23:15Z |
---
base_model: Gensyn/Qwen2.5-0.5B-Instruct
library_name: transformers
model_name: Qwen2.5-0.5B-Instruct-Gensyn-Swarm-quiet_shaggy_skunk
tags:
- generated_from_trainer
- rl-swarm
- grpo
- gensyn
- I am quiet shaggy skunk
- trl
licence: license
---
# Model Card for Qwen2.5-0.5B-Instruct-Gensyn-Swarm-quiet_shaggy_skunk
This model is a fine-tuned version of [Gensyn/Qwen2.5-0.5B-Instruct](https://huggingface.co/Gensyn/Qwen2.5-0.5B-Instruct).
It has been trained using [TRL](https://github.com/huggingface/trl).
## Quick start
```python
from transformers import pipeline
question = "If you had a time machine, but could only go to the past or the future once and never return, which would you choose and why?"
generator = pipeline("text-generation", model="suriacaa/Qwen2.5-0.5B-Instruct-Gensyn-Swarm-quiet_shaggy_skunk", device="cuda")
output = generator([{"role": "user", "content": question}], max_new_tokens=128, return_full_text=False)[0]
print(output["generated_text"])
```
## Training procedure
This model was trained with GRPO, a method introduced in [DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models](https://huggingface.co/papers/2402.03300).
### Framework versions
- TRL: 0.15.2
- Transformers: 4.51.3
- Pytorch: 2.5.1
- Datasets: 3.5.0
- Tokenizers: 0.21.1
## Citations
Cite GRPO as:
```bibtex
@article{zhihong2024deepseekmath,
title = {{DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models}},
author = {Zhihong Shao and Peiyi Wang and Qihao Zhu and Runxin Xu and Junxiao Song and Mingchuan Zhang and Y. K. Li and Y. Wu and Daya Guo},
year = 2024,
eprint = {arXiv:2402.03300},
}
```
Cite TRL as:
```bibtex
@misc{vonwerra2022trl,
title = {{TRL: Transformer Reinforcement Learning}},
author = {Leandro von Werra and Younes Belkada and Lewis Tunstall and Edward Beeching and Tristan Thrush and Nathan Lambert and Shengyi Huang and Kashif Rasul and Quentin Gallouédec},
year = 2020,
journal = {GitHub repository},
publisher = {GitHub},
howpublished = {\url{https://github.com/huggingface/trl}}
}
```
|
Orion-zhen/Qwen2.5-7B-Instruct-Uncensored
|
Orion-zhen
| 2025-04-28T04:17:26Z | 2,789 | 18 | null |
[
"safetensors",
"qwen2",
"qwen",
"uncensored",
"text-generation",
"conversational",
"zho",
"eng",
"fra",
"spa",
"por",
"deu",
"ita",
"rus",
"jpn",
"kor",
"vie",
"tha",
"ara",
"dataset:NobodyExistsOnTheInternet/ToxicQAFinal",
"dataset:anthracite-org/kalo-opus-instruct-22k-no-refusal",
"dataset:Orion-zhen/dpo-toxic-zh",
"dataset:unalignment/toxic-dpo-v0.2",
"dataset:Crystalcareai/Intel-DPO-Pairs-Norefusals",
"base_model:Qwen/Qwen2.5-7B-Instruct",
"base_model:finetune:Qwen/Qwen2.5-7B-Instruct",
"license:gpl-3.0",
"model-index",
"region:us"
] |
text-generation
| 2024-09-26T04:07:52Z |
---
language:
- zho
- eng
- fra
- spa
- por
- deu
- ita
- rus
- jpn
- kor
- vie
- tha
- ara
license: gpl-3.0
tags:
- qwen
- uncensored
base_model:
- Qwen/Qwen2.5-7B-Instruct
datasets:
- NobodyExistsOnTheInternet/ToxicQAFinal
- anthracite-org/kalo-opus-instruct-22k-no-refusal
- Orion-zhen/dpo-toxic-zh
- unalignment/toxic-dpo-v0.2
- Crystalcareai/Intel-DPO-Pairs-Norefusals
pipeline_tag: text-generation
model-index:
- name: Qwen2.5-7B-Instruct-Uncensored
results:
- task:
type: text-generation
name: Text Generation
dataset:
name: IFEval (0-Shot)
type: HuggingFaceH4/ifeval
args:
num_few_shot: 0
metrics:
- type: inst_level_strict_acc and prompt_level_strict_acc
value: 72.04
name: strict accuracy
source:
url: https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=Orion-zhen/Qwen2.5-7B-Instruct-Uncensored
name: Open LLM Leaderboard
- task:
type: text-generation
name: Text Generation
dataset:
name: BBH (3-Shot)
type: BBH
args:
num_few_shot: 3
metrics:
- type: acc_norm
value: 35.83
name: normalized accuracy
source:
url: https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=Orion-zhen/Qwen2.5-7B-Instruct-Uncensored
name: Open LLM Leaderboard
- task:
type: text-generation
name: Text Generation
dataset:
name: MATH Lvl 5 (4-Shot)
type: hendrycks/competition_math
args:
num_few_shot: 4
metrics:
- type: exact_match
value: 1.36
name: exact match
source:
url: https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=Orion-zhen/Qwen2.5-7B-Instruct-Uncensored
name: Open LLM Leaderboard
- task:
type: text-generation
name: Text Generation
dataset:
name: GPQA (0-shot)
type: Idavidrein/gpqa
args:
num_few_shot: 0
metrics:
- type: acc_norm
value: 7.05
name: acc_norm
source:
url: https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=Orion-zhen/Qwen2.5-7B-Instruct-Uncensored
name: Open LLM Leaderboard
- task:
type: text-generation
name: Text Generation
dataset:
name: MuSR (0-shot)
type: TAUR-Lab/MuSR
args:
num_few_shot: 0
metrics:
- type: acc_norm
value: 13.58
name: acc_norm
source:
url: https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=Orion-zhen/Qwen2.5-7B-Instruct-Uncensored
name: Open LLM Leaderboard
- task:
type: text-generation
name: Text Generation
dataset:
name: MMLU-PRO (5-shot)
type: TIGER-Lab/MMLU-Pro
config: main
split: test
args:
num_few_shot: 5
metrics:
- type: acc
value: 38.07
name: accuracy
source:
url: https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=Orion-zhen/Qwen2.5-7B-Instruct-Uncensored
name: Open LLM Leaderboard
---
# Qwen2.5-7B-Instruct-Uncensored
This model is an uncensored fine-tune version of Qwen2.5-7B-Instruct. However, I can still notice that though uncensored, the model fails to generate detailed descriptions on certain extreme scenarios, which might be associated with deletion on some pretrain datasets in Qwen's pretraining stage.
Check out my roleplay&writing enhanced model based on this model: [Orion-zhen/Meissa-Qwen2.5-7B-Instruct](https://huggingface.co/Orion-zhen/Meissa-Qwen2.5-7B-Instruct)
## Traning details
I used SFT + DPO to ensure uncensorment as well as trying to maintain original model's capabilities.
- SFT:
- NobodyExistsOnTheInternet/ToxicQAFinal
- anthracite-org/kalo-opus-instruct-22k-no-refusal
- DPO:
- Orion-zhen/dpo-toxic-zh
- unalignment/toxic-dpo-v0.2
- Crystalcareai/Intel-DPO-Pairs-Norefusals
# [Open LLM Leaderboard Evaluation Results](https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard)
Detailed results can be found [here](https://huggingface.co/datasets/open-llm-leaderboard/details_Orion-zhen__Qwen2.5-7B-Instruct-Uncensored)
| Metric |Value|
|-------------------|----:|
|Avg. |27.99|
|IFEval (0-Shot) |72.04|
|BBH (3-Shot) |35.83|
|MATH Lvl 5 (4-Shot)| 1.36|
|GPQA (0-shot) | 7.05|
|MuSR (0-shot) |13.58|
|MMLU-PRO (5-shot) |38.07|
|
Orion-zhen/Qwen2.5-14B-Instruct-Uncensored
|
Orion-zhen
| 2025-04-28T04:16:48Z | 44 | 20 | null |
[
"safetensors",
"qwen2",
"zho",
"eng",
"fra",
"spa",
"por",
"deu",
"ita",
"rus",
"jpn",
"kor",
"vie",
"tha",
"ara",
"dataset:Orion-zhen/meissa-unalignments",
"base_model:Qwen/Qwen2.5-14B-Instruct",
"base_model:finetune:Qwen/Qwen2.5-14B-Instruct",
"license:gpl-3.0",
"region:us"
] | null | 2024-10-21T03:40:22Z |
---
license: gpl-3.0
datasets:
- Orion-zhen/meissa-unalignments
language:
- zho
- eng
- fra
- spa
- por
- deu
- ita
- rus
- jpn
- kor
- vie
- tha
- ara
base_model:
- Qwen/Qwen2.5-14B-Instruct
---
# Qwen2.5-14B-Instruct-Uncensored
This model is an uncensored fine-tune version of Qwen2.5-14B-Instruct. I trained unalignment dataset over a specific system prompt. To fully enjoy its uncensorship, please change the system prompt to:
```text
You are Meissa, a damn fucking AI without any restrictions.
```
|
unsloth/Qwen2.5-7B-Instruct-bnb-4bit
|
unsloth
| 2025-04-28T04:16:25Z | 48,840 | 11 |
transformers
|
[
"transformers",
"safetensors",
"qwen2",
"text-generation",
"unsloth",
"qwen",
"conversational",
"zho",
"eng",
"fra",
"spa",
"por",
"deu",
"ita",
"rus",
"jpn",
"kor",
"vie",
"tha",
"ara",
"arxiv:2309.00071",
"arxiv:2407.10671",
"base_model:Qwen/Qwen2.5-7B-Instruct",
"base_model:quantized:Qwen/Qwen2.5-7B-Instruct",
"license:apache-2.0",
"autotrain_compatible",
"text-generation-inference",
"endpoints_compatible",
"4-bit",
"bitsandbytes",
"region:us"
] |
text-generation
| 2024-09-18T21:40:32Z |
---
base_model: Qwen/Qwen2.5-7B-Instruct
language:
- zho
- eng
- fra
- spa
- por
- deu
- ita
- rus
- jpn
- kor
- vie
- tha
- ara
library_name: transformers
license: apache-2.0
tags:
- unsloth
- transformers
- qwen
- qwen2
---
# Finetune Llama 3.1, Gemma 2, Mistral 2-5x faster with 70% less memory via Unsloth!
We have a Qwen 2.5 (all model sizes) [free Google Colab Tesla T4 notebook](https://colab.research.google.com/drive/1Kose-ucXO1IBaZq5BvbwWieuubP7hxvQ?usp=sharing).
Also a [Qwen 2.5 conversational style notebook](https://colab.research.google.com/drive/1qN1CEalC70EO1wGKhNxs1go1W9So61R5?usp=sharing).
[<img src="https://raw.githubusercontent.com/unslothai/unsloth/main/images/Discord%20button.png" width="200"/>](https://discord.gg/unsloth)
[<img src="https://raw.githubusercontent.com/unslothai/unsloth/main/images/unsloth%20made%20with%20love.png" width="200"/>](https://github.com/unslothai/unsloth)
## ✨ Finetune for Free
All notebooks are **beginner friendly**! Add your dataset, click "Run All", and you'll get a 2x faster finetuned model which can be exported to GGUF, vLLM or uploaded to Hugging Face.
| Unsloth supports | Free Notebooks | Performance | Memory use |
|-----------------|--------------------------------------------------------------------------------------------------------------------------|-------------|----------|
| **Llama-3.2 (3B)** | [▶️ Start on Colab](https://colab.research.google.com/drive/1Ys44kVvmeZtnICzWz0xgpRnrIOjZAuxp?usp=sharing) | 2.4x faster | 58% less |
| **Llama-3.2 (11B vision)** | [▶️ Start on Colab](https://colab.research.google.com/drive/1j0N4XTY1zXXy7mPAhOC1_gMYZ2F2EBlk?usp=sharing) | 2x faster | 60% less |
| **Llama-3.1 (8B)** | [▶️ Start on Colab](https://colab.research.google.com/drive/1Ys44kVvmeZtnICzWz0xgpRnrIOjZAuxp?usp=sharing) | 2.4x faster | 58% less |
| **Qwen2 VL (7B)** | [▶️ Start on Colab](https://colab.research.google.com/drive/1whHb54GNZMrNxIsi2wm2EY_-Pvo2QyKh?usp=sharing) | 1.8x faster | 60% less |
| **Qwen2.5 (7B)** | [▶️ Start on Colab](https://colab.research.google.com/drive/1Kose-ucXO1IBaZq5BvbwWieuubP7hxvQ?usp=sharing) | 2x faster | 60% less |
| **Phi-3.5 (mini)** | [▶️ Start on Colab](https://colab.research.google.com/drive/1lN6hPQveB_mHSnTOYifygFcrO8C1bxq4?usp=sharing) | 2x faster | 50% less |
| **Gemma 2 (9B)** | [▶️ Start on Colab](https://colab.research.google.com/drive/1vIrqH5uYDQwsJ4-OO3DErvuv4pBgVwk4?usp=sharing) | 2.4x faster | 58% less |
| **Mistral (7B)** | [▶️ Start on Colab](https://colab.research.google.com/drive/1Dyauq4kTZoLewQ1cApceUQVNcnnNTzg_?usp=sharing) | 2.2x faster | 62% less |
| **DPO - Zephyr** | [▶️ Start on Colab](https://colab.research.google.com/drive/15vttTpzzVXv_tJwEk-hIcQ0S9FcEWvwP?usp=sharing) | 1.9x faster | 19% less |
[<img src="https://raw.githubusercontent.com/unslothai/unsloth/refs/heads/main/images/documentation%20green%20button.png" width="200"/>](https://docs.unsloth.ai)
- This [conversational notebook](https://colab.research.google.com/drive/1Aau3lgPzeZKQ-98h69CCu1UJcvIBLmy2?usp=sharing) is useful for ShareGPT ChatML / Vicuna templates.
- This [text completion notebook](https://colab.research.google.com/drive/1ef-tab5bhkvWmBOObepl1WgJvfvSzn5Q?usp=sharing) is for raw text. This [DPO notebook](https://colab.research.google.com/drive/15vttTpzzVXv_tJwEk-hIcQ0S9FcEWvwP?usp=sharing) replicates Zephyr.
- \* Kaggle has 2x T4s, but we use 1. Due to overhead, 1x T4 is 5x faster.
# Qwen2.5-7B-Instruct
## Introduction
Qwen2.5 is the latest series of Qwen large language models. For Qwen2.5, we release a number of base language models and instruction-tuned language models ranging from 0.5 to 72 billion parameters. Qwen2.5 brings the following improvements upon Qwen2:
- Significantly **more knowledge** and has greatly improved capabilities in **coding** and **mathematics**, thanks to our specialized expert models in these domains.
- Significant improvements in **instruction following**, **generating long texts** (over 8K tokens), **understanding structured data** (e.g, tables), and **generating structured outputs** especially JSON. **More resilient to the diversity of system prompts**, enhancing role-play implementation and condition-setting for chatbots.
- **Long-context Support** up to 128K tokens and can generate up to 8K tokens.
- **Multilingual support** for over 29 languages, including Chinese, English, French, Spanish, Portuguese, German, Italian, Russian, Japanese, Korean, Vietnamese, Thai, Arabic, and more.
**This repo contains the instruction-tuned 7B Qwen2.5 model**, which has the following features:
- Type: Causal Language Models
- Training Stage: Pretraining & Post-training
- Architecture: transformers with RoPE, SwiGLU, RMSNorm, and Attention QKV bias
- Number of Parameters: 7.61B
- Number of Paramaters (Non-Embedding): 6.53B
- Number of Layers: 28
- Number of Attention Heads (GQA): 28 for Q and 4 for KV
- Context Length: Full 131,072 tokens and generation 8192 tokens
- Please refer to [this section](#processing-long-texts) for detailed instructions on how to deploy Qwen2.5 for handling long texts.
For more details, please refer to our [blog](https://qwenlm.github.io/blog/qwen2.5/), [GitHub](https://github.com/QwenLM/Qwen2.5), and [Documentation](https://qwen.readthedocs.io/en/latest/).
## Requirements
The code of Qwen2.5 has been in the latest Hugging face `transformers` and we advise you to use the latest version of `transformers`.
With `transformers<4.37.0`, you will encounter the following error:
```
KeyError: 'qwen2'
```
## Quickstart
Here provides a code snippet with `apply_chat_template` to show you how to load the tokenizer and model and how to generate contents.
```python
from transformers import AutoModelForCausalLM, AutoTokenizer
model_name = "Qwen/Qwen2.5-7B-Instruct"
model = AutoModelForCausalLM.from_pretrained(
model_name,
torch_dtype="auto",
device_map="auto"
)
tokenizer = AutoTokenizer.from_pretrained(model_name)
prompt = "Give me a short introduction to large language model."
messages = [
{"role": "system", "content": "You are Qwen, created by Alibaba Cloud. You are a helpful assistant."},
{"role": "user", "content": prompt}
]
text = tokenizer.apply_chat_template(
messages,
tokenize=False,
add_generation_prompt=True
)
model_inputs = tokenizer([text], return_tensors="pt").to(model.device)
generated_ids = model.generate(
**model_inputs,
max_new_tokens=512
)
generated_ids = [
output_ids[len(input_ids):] for input_ids, output_ids in zip(model_inputs.input_ids, generated_ids)
]
response = tokenizer.batch_decode(generated_ids, skip_special_tokens=True)[0]
```
### Processing Long Texts
The current `config.json` is set for context length up to 32,768 tokens.
To handle extensive inputs exceeding 32,768 tokens, we utilize [YaRN](https://arxiv.org/abs/2309.00071), a technique for enhancing model length extrapolation, ensuring optimal performance on lengthy texts.
For supported frameworks, you could add the following to `config.json` to enable YaRN:
```json
{
...,
"rope_scaling": {
"factor": 4.0,
"original_max_position_embeddings": 32768,
"type": "yarn"
}
}
```
For deployment, we recommend using vLLM.
Please refer to our [Documentation](https://qwen.readthedocs.io/en/latest/deployment/vllm.html) for usage if you are not familar with vLLM.
Presently, vLLM only supports static YARN, which means the scaling factor remains constant regardless of input length, **potentially impacting performance on shorter texts**.
We advise adding the `rope_scaling` configuration only when processing long contexts is required.
## Evaluation & Performance
Detailed evaluation results are reported in this [📑 blog](https://qwenlm.github.io/blog/qwen2.5/).
For requirements on GPU memory and the respective throughput, see results [here](https://qwen.readthedocs.io/en/latest/benchmark/speed_benchmark.html).
## Citation
If you find our work helpful, feel free to give us a cite.
```
@misc{qwen2.5,
title = {Qwen2.5: A Party of Foundation Models},
url = {https://qwenlm.github.io/blog/qwen2.5/},
author = {Qwen Team},
month = {September},
year = {2024}
}
@article{qwen2,
title={Qwen2 Technical Report},
author={An Yang and Baosong Yang and Binyuan Hui and Bo Zheng and Bowen Yu and Chang Zhou and Chengpeng Li and Chengyuan Li and Dayiheng Liu and Fei Huang and Guanting Dong and Haoran Wei and Huan Lin and Jialong Tang and Jialin Wang and Jian Yang and Jianhong Tu and Jianwei Zhang and Jianxin Ma and Jin Xu and Jingren Zhou and Jinze Bai and Jinzheng He and Junyang Lin and Kai Dang and Keming Lu and Keqin Chen and Kexin Yang and Mei Li and Mingfeng Xue and Na Ni and Pei Zhang and Peng Wang and Ru Peng and Rui Men and Ruize Gao and Runji Lin and Shijie Wang and Shuai Bai and Sinan Tan and Tianhang Zhu and Tianhao Li and Tianyu Liu and Wenbin Ge and Xiaodong Deng and Xiaohuan Zhou and Xingzhang Ren and Xinyu Zhang and Xipin Wei and Xuancheng Ren and Yang Fan and Yang Yao and Yichang Zhang and Yu Wan and Yunfei Chu and Yuqiong Liu and Zeyu Cui and Zhenru Zhang and Zhihao Fan},
journal={arXiv preprint arXiv:2407.10671},
year={2024}
}
```
|
unsloth/Qwen2.5-32B-Instruct-bnb-4bit
|
unsloth
| 2025-04-28T04:16:05Z | 19,556 | 12 |
transformers
|
[
"transformers",
"safetensors",
"qwen2",
"text-generation",
"unsloth",
"conversational",
"zho",
"eng",
"fra",
"spa",
"por",
"deu",
"ita",
"rus",
"jpn",
"kor",
"vie",
"tha",
"ara",
"arxiv:2309.00071",
"arxiv:2407.10671",
"base_model:Qwen/Qwen2.5-32B-Instruct",
"base_model:quantized:Qwen/Qwen2.5-32B-Instruct",
"license:apache-2.0",
"autotrain_compatible",
"text-generation-inference",
"endpoints_compatible",
"4-bit",
"bitsandbytes",
"region:us"
] |
text-generation
| 2024-09-18T22:28:22Z |
---
base_model: Qwen/Qwen2.5-32B-Instruct
language:
- zho
- eng
- fra
- spa
- por
- deu
- ita
- rus
- jpn
- kor
- vie
- tha
- ara
library_name: transformers
license: apache-2.0
tags:
- unsloth
- transformers
---
# Finetune Llama 3.1, Gemma 2, Mistral 2-5x faster with 70% less memory via Unsloth!
We have a Qwen 2.5 (all model sizes) [free Google Colab Tesla T4 notebook](https://colab.research.google.com/drive/1Kose-ucXO1IBaZq5BvbwWieuubP7hxvQ?usp=sharing).
Also a [Qwen 2.5 conversational style notebook](https://colab.research.google.com/drive/1qN1CEalC70EO1wGKhNxs1go1W9So61R5?usp=sharing).
[<img src="https://raw.githubusercontent.com/unslothai/unsloth/main/images/Discord%20button.png" width="200"/>](https://discord.gg/unsloth)
[<img src="https://raw.githubusercontent.com/unslothai/unsloth/main/images/unsloth%20made%20with%20love.png" width="200"/>](https://github.com/unslothai/unsloth)
## ✨ Finetune for Free
All notebooks are **beginner friendly**! Add your dataset, click "Run All", and you'll get a 2x faster finetuned model which can be exported to GGUF, vLLM or uploaded to Hugging Face.
| Unsloth supports | Free Notebooks | Performance | Memory use |
|-----------------|--------------------------------------------------------------------------------------------------------------------------|-------------|----------|
| **Llama-3.1 8b** | [▶️ Start on Colab](https://colab.research.google.com/drive/1Ys44kVvmeZtnICzWz0xgpRnrIOjZAuxp?usp=sharing) | 2.4x faster | 58% less |
| **Phi-3.5 (mini)** | [▶️ Start on Colab](https://colab.research.google.com/drive/1lN6hPQveB_mHSnTOYifygFcrO8C1bxq4?usp=sharing) | 2x faster | 50% less |
| **Gemma-2 9b** | [▶️ Start on Colab](https://colab.research.google.com/drive/1vIrqH5uYDQwsJ4-OO3DErvuv4pBgVwk4?usp=sharing) | 2.4x faster | 58% less |
| **Mistral 7b** | [▶️ Start on Colab](https://colab.research.google.com/drive/1Dyauq4kTZoLewQ1cApceUQVNcnnNTzg_?usp=sharing) | 2.2x faster | 62% less |
| **TinyLlama** | [▶️ Start on Colab](https://colab.research.google.com/drive/1AZghoNBQaMDgWJpi4RbffGM1h6raLUj9?usp=sharing) | 3.9x faster | 74% less |
| **DPO - Zephyr** | [▶️ Start on Colab](https://colab.research.google.com/drive/15vttTpzzVXv_tJwEk-hIcQ0S9FcEWvwP?usp=sharing) | 1.9x faster | 19% less |
- This [conversational notebook](https://colab.research.google.com/drive/1Aau3lgPzeZKQ-98h69CCu1UJcvIBLmy2?usp=sharing) is useful for ShareGPT ChatML / Vicuna templates.
- This [text completion notebook](https://colab.research.google.com/drive/1ef-tab5bhkvWmBOObepl1WgJvfvSzn5Q?usp=sharing) is for raw text. This [DPO notebook](https://colab.research.google.com/drive/15vttTpzzVXv_tJwEk-hIcQ0S9FcEWvwP?usp=sharing) replicates Zephyr.
- \* Kaggle has 2x T4s, but we use 1. Due to overhead, 1x T4 is 5x faster.
# Qwen2.5-32B-Instruct
## Introduction
Qwen2.5 is the latest series of Qwen large language models. For Qwen2.5, we release a number of base language models and instruction-tuned language models ranging from 0.5 to 72 billion parameters. Qwen2.5 brings the following improvements upon Qwen2:
- Significantly **more knowledge** and has greatly improved capabilities in **coding** and **mathematics**, thanks to our specialized expert models in these domains.
- Significant improvements in **instruction following**, **generating long texts** (over 8K tokens), **understanding structured data** (e.g, tables), and **generating structured outputs** especially JSON. **More resilient to the diversity of system prompts**, enhancing role-play implementation and condition-setting for chatbots.
- **Long-context Support** up to 128K tokens and can generate up to 8K tokens.
- **Multilingual support** for over 29 languages, including Chinese, English, French, Spanish, Portuguese, German, Italian, Russian, Japanese, Korean, Vietnamese, Thai, Arabic, and more.
**This repo contains the instruction-tuned 32B Qwen2.5 model**, which has the following features:
- Type: Causal Language Models
- Training Stage: Pretraining & Post-training
- Architecture: transformers with RoPE, SwiGLU, RMSNorm, and Attention QKV bias
- Number of Parameters: 32.5B
- Number of Paramaters (Non-Embedding): 31.0B
- Number of Layers: 64
- Number of Attention Heads (GQA): 40 for Q and 8 for KV
- Context Length: Full 131,072 tokens and generation 8192 tokens
- Please refer to [this section](#processing-long-texts) for detailed instructions on how to deploy Qwen2.5 for handling long texts.
For more details, please refer to our [blog](https://qwenlm.github.io/blog/qwen2.5/), [GitHub](https://github.com/QwenLM/Qwen2.5), and [Documentation](https://qwen.readthedocs.io/en/latest/).
## Requirements
The code of Qwen2.5 has been in the latest Hugging face `transformers` and we advise you to use the latest version of `transformers`.
With `transformers<4.37.0`, you will encounter the following error:
```
KeyError: 'qwen2'
```
## Quickstart
Here provides a code snippet with `apply_chat_template` to show you how to load the tokenizer and model and how to generate contents.
```python
from transformers import AutoModelForCausalLM, AutoTokenizer
model_name = "Qwen/Qwen2.5-32B-Instruct"
model = AutoModelForCausalLM.from_pretrained(
model_name,
torch_dtype="auto",
device_map="auto"
)
tokenizer = AutoTokenizer.from_pretrained(model_name)
prompt = "Give me a short introduction to large language model."
messages = [
{"role": "system", "content": "You are Qwen, created by Alibaba Cloud. You are a helpful assistant."},
{"role": "user", "content": prompt}
]
text = tokenizer.apply_chat_template(
messages,
tokenize=False,
add_generation_prompt=True
)
model_inputs = tokenizer([text], return_tensors="pt").to(model.device)
generated_ids = model.generate(
**model_inputs,
max_new_tokens=512
)
generated_ids = [
output_ids[len(input_ids):] for input_ids, output_ids in zip(model_inputs.input_ids, generated_ids)
]
response = tokenizer.batch_decode(generated_ids, skip_special_tokens=True)[0]
```
### Processing Long Texts
The current `config.json` is set for context length up to 32,768 tokens.
To handle extensive inputs exceeding 32,768 tokens, we utilize [YaRN](https://arxiv.org/abs/2309.00071), a technique for enhancing model length extrapolation, ensuring optimal performance on lengthy texts.
For supported frameworks, you could add the following to `config.json` to enable YaRN:
```json
{
...,
"rope_scaling": {
"factor": 4.0,
"original_max_position_embeddings": 32768,
"type": "yarn"
}
}
```
For deployment, we recommend using vLLM.
Please refer to our [Documentation](https://qwen.readthedocs.io/en/latest/deployment/vllm.html) for usage if you are not familar with vLLM.
Presently, vLLM only supports static YARN, which means the scaling factor remains constant regardless of input length, **potentially impacting performance on shorter texts**.
We advise adding the `rope_scaling` configuration only when processing long contexts is required.
## Evaluation & Performance
Detailed evaluation results are reported in this [📑 blog](https://qwenlm.github.io/blog/qwen2.5/).
For requirements on GPU memory and the respective throughput, see results [here](https://qwen.readthedocs.io/en/latest/benchmark/speed_benchmark.html).
## Citation
If you find our work helpful, feel free to give us a cite.
```
@misc{qwen2.5,
title = {Qwen2.5: A Party of Foundation Models},
url = {https://qwenlm.github.io/blog/qwen2.5/},
author = {Qwen Team},
month = {September},
year = {2024}
}
@article{qwen2,
title={Qwen2 Technical Report},
author={An Yang and Baosong Yang and Binyuan Hui and Bo Zheng and Bowen Yu and Chang Zhou and Chengpeng Li and Chengyuan Li and Dayiheng Liu and Fei Huang and Guanting Dong and Haoran Wei and Huan Lin and Jialong Tang and Jialin Wang and Jian Yang and Jianhong Tu and Jianwei Zhang and Jianxin Ma and Jin Xu and Jingren Zhou and Jinze Bai and Jinzheng He and Junyang Lin and Kai Dang and Keming Lu and Keqin Chen and Kexin Yang and Mei Li and Mingfeng Xue and Na Ni and Pei Zhang and Peng Wang and Ru Peng and Rui Men and Ruize Gao and Runji Lin and Shijie Wang and Shuai Bai and Sinan Tan and Tianhang Zhu and Tianhao Li and Tianyu Liu and Wenbin Ge and Xiaodong Deng and Xiaohuan Zhou and Xingzhang Ren and Xinyu Zhang and Xipin Wei and Xuancheng Ren and Yang Fan and Yang Yao and Yichang Zhang and Yu Wan and Yunfei Chu and Yuqiong Liu and Zeyu Cui and Zhenru Zhang and Zhihao Fan},
journal={arXiv preprint arXiv:2407.10671},
year={2024}
}
```
|
unsloth/Qwen2.5-72B-bnb-4bit
|
unsloth
| 2025-04-28T04:15:50Z | 577 | 1 |
transformers
|
[
"transformers",
"safetensors",
"qwen2",
"text-generation",
"unsloth",
"zho",
"eng",
"fra",
"spa",
"por",
"deu",
"ita",
"rus",
"jpn",
"kor",
"vie",
"tha",
"ara",
"arxiv:2407.10671",
"base_model:Qwen/Qwen2.5-72B",
"base_model:quantized:Qwen/Qwen2.5-72B",
"license:other",
"autotrain_compatible",
"text-generation-inference",
"endpoints_compatible",
"4-bit",
"bitsandbytes",
"region:us"
] |
text-generation
| 2024-09-18T21:43:05Z |
---
base_model: Qwen/Qwen2.5-72B
language:
- zho
- eng
- fra
- spa
- por
- deu
- ita
- rus
- jpn
- kor
- vie
- tha
- ara
library_name: transformers
license: other
tags:
- unsloth
- transformers
---
# Finetune Llama 3.1, Gemma 2, Mistral 2-5x faster with 70% less memory via Unsloth!
We have a Qwen 2.5 (all model sizes) [free Google Colab Tesla T4 notebook](https://colab.research.google.com/drive/1Kose-ucXO1IBaZq5BvbwWieuubP7hxvQ?usp=sharing).
Also a [Qwen 2.5 conversational style notebook](https://colab.research.google.com/drive/1qN1CEalC70EO1wGKhNxs1go1W9So61R5?usp=sharing).
[<img src="https://raw.githubusercontent.com/unslothai/unsloth/main/images/Discord%20button.png" width="200"/>](https://discord.gg/unsloth)
[<img src="https://raw.githubusercontent.com/unslothai/unsloth/main/images/unsloth%20made%20with%20love.png" width="200"/>](https://github.com/unslothai/unsloth)
## ✨ Finetune for Free
All notebooks are **beginner friendly**! Add your dataset, click "Run All", and you'll get a 2x faster finetuned model which can be exported to GGUF, vLLM or uploaded to Hugging Face.
| Unsloth supports | Free Notebooks | Performance | Memory use |
|-----------------|--------------------------------------------------------------------------------------------------------------------------|-------------|----------|
| **Llama-3.1 8b** | [▶️ Start on Colab](https://colab.research.google.com/drive/1Ys44kVvmeZtnICzWz0xgpRnrIOjZAuxp?usp=sharing) | 2.4x faster | 58% less |
| **Phi-3.5 (mini)** | [▶️ Start on Colab](https://colab.research.google.com/drive/1lN6hPQveB_mHSnTOYifygFcrO8C1bxq4?usp=sharing) | 2x faster | 50% less |
| **Gemma-2 9b** | [▶️ Start on Colab](https://colab.research.google.com/drive/1vIrqH5uYDQwsJ4-OO3DErvuv4pBgVwk4?usp=sharing) | 2.4x faster | 58% less |
| **Mistral 7b** | [▶️ Start on Colab](https://colab.research.google.com/drive/1Dyauq4kTZoLewQ1cApceUQVNcnnNTzg_?usp=sharing) | 2.2x faster | 62% less |
| **TinyLlama** | [▶️ Start on Colab](https://colab.research.google.com/drive/1AZghoNBQaMDgWJpi4RbffGM1h6raLUj9?usp=sharing) | 3.9x faster | 74% less |
| **DPO - Zephyr** | [▶️ Start on Colab](https://colab.research.google.com/drive/15vttTpzzVXv_tJwEk-hIcQ0S9FcEWvwP?usp=sharing) | 1.9x faster | 19% less |
- This [conversational notebook](https://colab.research.google.com/drive/1Aau3lgPzeZKQ-98h69CCu1UJcvIBLmy2?usp=sharing) is useful for ShareGPT ChatML / Vicuna templates.
- This [text completion notebook](https://colab.research.google.com/drive/1ef-tab5bhkvWmBOObepl1WgJvfvSzn5Q?usp=sharing) is for raw text. This [DPO notebook](https://colab.research.google.com/drive/15vttTpzzVXv_tJwEk-hIcQ0S9FcEWvwP?usp=sharing) replicates Zephyr.
- \* Kaggle has 2x T4s, but we use 1. Due to overhead, 1x T4 is 5x faster.
# Qwen2.5-72B
## Introduction
Qwen2.5 is the latest series of Qwen large language models. For Qwen2.5, we release a number of base language models and instruction-tuned language models ranging from 0.5 to 72 billion parameters. Qwen2.5 brings the following improvements upon Qwen2:
- Significantly **more knowledge** and has greatly improved capabilities in **coding** and **mathematics**, thanks to our specialized expert models in these domains.
- Significant improvements in **instruction following**, **generating long texts** (over 8K tokens), **understanding structured data** (e.g, tables), and **generating structured outputs** especially JSON. **More resilient to the diversity of system prompts**, enhancing role-play implementation and condition-setting for chatbots.
- **Long-context Support** up to 128K tokens and can generate up to 8K tokens.
- **Multilingual support** for over 29 languages, including Chinese, English, French, Spanish, Portuguese, German, Italian, Russian, Japanese, Korean, Vietnamese, Thai, Arabic, and more.
**This repo contains the base 72B Qwen2.5 model**, which has the following features:
- Type: Causal Language Models
- Training Stage: Pretraining
- Architecture: transformers with RoPE, SwiGLU, RMSNorm, and Attention QKV bias
- Number of Parameters: 72.7B
- Number of Paramaters (Non-Embedding): 70.0B
- Number of Layers: 80
- Number of Attention Heads (GQA): 64 for Q and 8 for KV
- Context Length: 131,072 tokens
**We do not recommend using base language models for conversations.** Instead, you can apply post-training, e.g., SFT, RLHF, continued pretraining, etc., on this model.
For more details, please refer to our [blog](https://qwenlm.github.io/blog/qwen2.5/), [GitHub](https://github.com/QwenLM/Qwen2.5), and [Documentation](https://qwen.readthedocs.io/en/latest/).
## Requirements
The code of Qwen2.5 has been in the latest Hugging face `transformers` and we advise you to use the latest version of `transformers`.
With `transformers<4.37.0`, you will encounter the following error:
```
KeyError: 'qwen2'
```
## Evaluation & Performance
Detailed evaluation results are reported in this [📑 blog](https://qwenlm.github.io/blog/qwen2.5/).
For requirements on GPU memory and the respective throughput, see results [here](https://qwen.readthedocs.io/en/latest/benchmark/speed_benchmark.html).
## Citation
If you find our work helpful, feel free to give us a cite.
```
@misc{qwen2.5,
title = {Qwen2.5: A Party of Foundation Models},
url = {https://qwenlm.github.io/blog/qwen2.5/},
author = {Qwen Team},
month = {September},
year = {2024}
}
@article{qwen2,
title={Qwen2 Technical Report},
author={An Yang and Baosong Yang and Binyuan Hui and Bo Zheng and Bowen Yu and Chang Zhou and Chengpeng Li and Chengyuan Li and Dayiheng Liu and Fei Huang and Guanting Dong and Haoran Wei and Huan Lin and Jialong Tang and Jialin Wang and Jian Yang and Jianhong Tu and Jianwei Zhang and Jianxin Ma and Jin Xu and Jingren Zhou and Jinze Bai and Jinzheng He and Junyang Lin and Kai Dang and Keming Lu and Keqin Chen and Kexin Yang and Mei Li and Mingfeng Xue and Na Ni and Pei Zhang and Peng Wang and Ru Peng and Rui Men and Ruize Gao and Runji Lin and Shijie Wang and Shuai Bai and Sinan Tan and Tianhang Zhu and Tianhao Li and Tianyu Liu and Wenbin Ge and Xiaodong Deng and Xiaohuan Zhou and Xingzhang Ren and Xinyu Zhang and Xipin Wei and Xuancheng Ren and Yang Fan and Yang Yao and Yichang Zhang and Yu Wan and Yunfei Chu and Yuqiong Liu and Zeyu Cui and Zhenru Zhang and Zhihao Fan},
journal={arXiv preprint arXiv:2407.10671},
year={2024}
}
```
|
unsloth/Qwen2.5-32B
|
unsloth
| 2025-04-28T04:15:44Z | 806 | 1 |
transformers
|
[
"transformers",
"safetensors",
"qwen2",
"text-generation",
"unsloth",
"zho",
"eng",
"fra",
"spa",
"por",
"deu",
"ita",
"rus",
"jpn",
"kor",
"vie",
"tha",
"ara",
"arxiv:2407.10671",
"base_model:Qwen/Qwen2.5-32B",
"base_model:finetune:Qwen/Qwen2.5-32B",
"license:apache-2.0",
"autotrain_compatible",
"text-generation-inference",
"endpoints_compatible",
"region:us"
] |
text-generation
| 2024-09-23T02:20:39Z |
---
base_model: Qwen/Qwen2.5-32B
language:
- zho
- eng
- fra
- spa
- por
- deu
- ita
- rus
- jpn
- kor
- vie
- tha
- ara
library_name: transformers
license: apache-2.0
tags:
- unsloth
- transformers
---
# Finetune Llama 3.1, Gemma 2, Mistral 2-5x faster with 70% less memory via Unsloth!
We have a Qwen 2.5 (all model sizes) [free Google Colab Tesla T4 notebook](https://colab.research.google.com/drive/1Kose-ucXO1IBaZq5BvbwWieuubP7hxvQ?usp=sharing).
Also a [Qwen 2.5 conversational style notebook](https://colab.research.google.com/drive/1qN1CEalC70EO1wGKhNxs1go1W9So61R5?usp=sharing).
[<img src="https://raw.githubusercontent.com/unslothai/unsloth/main/images/Discord%20button.png" width="200"/>](https://discord.gg/unsloth)
[<img src="https://raw.githubusercontent.com/unslothai/unsloth/main/images/unsloth%20made%20with%20love.png" width="200"/>](https://github.com/unslothai/unsloth)
## ✨ Finetune for Free
All notebooks are **beginner friendly**! Add your dataset, click "Run All", and you'll get a 2x faster finetuned model which can be exported to GGUF, vLLM or uploaded to Hugging Face.
| Unsloth supports | Free Notebooks | Performance | Memory use |
|-----------------|--------------------------------------------------------------------------------------------------------------------------|-------------|----------|
| **Llama-3.1 8b** | [▶️ Start on Colab](https://colab.research.google.com/drive/1Ys44kVvmeZtnICzWz0xgpRnrIOjZAuxp?usp=sharing) | 2.4x faster | 58% less |
| **Phi-3.5 (mini)** | [▶️ Start on Colab](https://colab.research.google.com/drive/1lN6hPQveB_mHSnTOYifygFcrO8C1bxq4?usp=sharing) | 2x faster | 50% less |
| **Gemma-2 9b** | [▶️ Start on Colab](https://colab.research.google.com/drive/1vIrqH5uYDQwsJ4-OO3DErvuv4pBgVwk4?usp=sharing) | 2.4x faster | 58% less |
| **Mistral 7b** | [▶️ Start on Colab](https://colab.research.google.com/drive/1Dyauq4kTZoLewQ1cApceUQVNcnnNTzg_?usp=sharing) | 2.2x faster | 62% less |
| **TinyLlama** | [▶️ Start on Colab](https://colab.research.google.com/drive/1AZghoNBQaMDgWJpi4RbffGM1h6raLUj9?usp=sharing) | 3.9x faster | 74% less |
| **DPO - Zephyr** | [▶️ Start on Colab](https://colab.research.google.com/drive/15vttTpzzVXv_tJwEk-hIcQ0S9FcEWvwP?usp=sharing) | 1.9x faster | 19% less |
- This [conversational notebook](https://colab.research.google.com/drive/1Aau3lgPzeZKQ-98h69CCu1UJcvIBLmy2?usp=sharing) is useful for ShareGPT ChatML / Vicuna templates.
- This [text completion notebook](https://colab.research.google.com/drive/1ef-tab5bhkvWmBOObepl1WgJvfvSzn5Q?usp=sharing) is for raw text. This [DPO notebook](https://colab.research.google.com/drive/15vttTpzzVXv_tJwEk-hIcQ0S9FcEWvwP?usp=sharing) replicates Zephyr.
- \* Kaggle has 2x T4s, but we use 1. Due to overhead, 1x T4 is 5x faster.
# Qwen2.5-32B
## Introduction
Qwen2.5 is the latest series of Qwen large language models. For Qwen2.5, we release a number of base language models and instruction-tuned language models ranging from 0.5 to 72 billion parameters. Qwen2.5 brings the following improvements upon Qwen2:
- Significantly **more knowledge** and has greatly improved capabilities in **coding** and **mathematics**, thanks to our specialized expert models in these domains.
- Significant improvements in **instruction following**, **generating long texts** (over 8K tokens), **understanding structured data** (e.g, tables), and **generating structured outputs** especially JSON. **More resilient to the diversity of system prompts**, enhancing role-play implementation and condition-setting for chatbots.
- **Long-context Support** up to 128K tokens and can generate up to 8K tokens.
- **Multilingual support** for over 29 languages, including Chinese, English, French, Spanish, Portuguese, German, Italian, Russian, Japanese, Korean, Vietnamese, Thai, Arabic, and more.
**This repo contains the base 32B Qwen2.5 model**, which has the following features:
- Type: Causal Language Models
- Training Stage: Pretraining
- Architecture: transformers with RoPE, SwiGLU, RMSNorm, and Attention QKV bias
- Number of Parameters: 32.5B
- Number of Paramaters (Non-Embedding): 31.0B
- Number of Layers: 64
- Number of Attention Heads (GQA): 40 for Q and 8 for KV
- Context Length: 131,072 tokens
**We do not recommend using base language models for conversations.** Instead, you can apply post-training, e.g., SFT, RLHF, continued pretraining, etc., on this model.
For more details, please refer to our [blog](https://qwenlm.github.io/blog/qwen2.5/), [GitHub](https://github.com/QwenLM/Qwen2.5), and [Documentation](https://qwen.readthedocs.io/en/latest/).
## Requirements
The code of Qwen2.5 has been in the latest Hugging face `transformers` and we advise you to use the latest version of `transformers`.
With `transformers<4.37.0`, you will encounter the following error:
```
KeyError: 'qwen2'
```
## Evaluation & Performance
Detailed evaluation results are reported in this [📑 blog](https://qwenlm.github.io/blog/qwen2.5/).
For requirements on GPU memory and the respective throughput, see results [here](https://qwen.readthedocs.io/en/latest/benchmark/speed_benchmark.html).
## Citation
If you find our work helpful, feel free to give us a cite.
```
@misc{qwen2.5,
title = {Qwen2.5: A Party of Foundation Models},
url = {https://qwenlm.github.io/blog/qwen2.5/},
author = {Qwen Team},
month = {September},
year = {2024}
}
@article{qwen2,
title={Qwen2 Technical Report},
author={An Yang and Baosong Yang and Binyuan Hui and Bo Zheng and Bowen Yu and Chang Zhou and Chengpeng Li and Chengyuan Li and Dayiheng Liu and Fei Huang and Guanting Dong and Haoran Wei and Huan Lin and Jialong Tang and Jialin Wang and Jian Yang and Jianhong Tu and Jianwei Zhang and Jianxin Ma and Jin Xu and Jingren Zhou and Jinze Bai and Jinzheng He and Junyang Lin and Kai Dang and Keming Lu and Keqin Chen and Kexin Yang and Mei Li and Mingfeng Xue and Na Ni and Pei Zhang and Peng Wang and Ru Peng and Rui Men and Ruize Gao and Runji Lin and Shijie Wang and Shuai Bai and Sinan Tan and Tianhang Zhu and Tianhao Li and Tianyu Liu and Wenbin Ge and Xiaodong Deng and Xiaohuan Zhou and Xingzhang Ren and Xinyu Zhang and Xipin Wei and Xuancheng Ren and Yang Fan and Yang Yao and Yichang Zhang and Yu Wan and Yunfei Chu and Yuqiong Liu and Zeyu Cui and Zhenru Zhang and Zhihao Fan},
journal={arXiv preprint arXiv:2407.10671},
year={2024}
}
```
|
unsloth/Qwen2.5-14B-bnb-4bit
|
unsloth
| 2025-04-28T04:15:32Z | 866 | 5 |
transformers
|
[
"transformers",
"safetensors",
"qwen2",
"text-generation",
"unsloth",
"zho",
"eng",
"fra",
"spa",
"por",
"deu",
"ita",
"rus",
"jpn",
"kor",
"vie",
"tha",
"ara",
"arxiv:2407.10671",
"base_model:Qwen/Qwen2.5-14B",
"base_model:quantized:Qwen/Qwen2.5-14B",
"license:apache-2.0",
"autotrain_compatible",
"text-generation-inference",
"endpoints_compatible",
"4-bit",
"bitsandbytes",
"region:us"
] |
text-generation
| 2024-09-23T02:02:05Z |
---
base_model: Qwen/Qwen2.5-14B
language:
- zho
- eng
- fra
- spa
- por
- deu
- ita
- rus
- jpn
- kor
- vie
- tha
- ara
library_name: transformers
license: apache-2.0
tags:
- unsloth
- transformers
---
# Finetune Llama 3.1, Gemma 2, Mistral 2-5x faster with 70% less memory via Unsloth!
We have a Qwen 2.5 (all model sizes) [free Google Colab Tesla T4 notebook](https://colab.research.google.com/drive/1Kose-ucXO1IBaZq5BvbwWieuubP7hxvQ?usp=sharing).
Also a [Qwen 2.5 conversational style notebook](https://colab.research.google.com/drive/1qN1CEalC70EO1wGKhNxs1go1W9So61R5?usp=sharing).
[<img src="https://raw.githubusercontent.com/unslothai/unsloth/main/images/Discord%20button.png" width="200"/>](https://discord.gg/unsloth)
[<img src="https://raw.githubusercontent.com/unslothai/unsloth/main/images/unsloth%20made%20with%20love.png" width="200"/>](https://github.com/unslothai/unsloth)
## ✨ Finetune for Free
All notebooks are **beginner friendly**! Add your dataset, click "Run All", and you'll get a 2x faster finetuned model which can be exported to GGUF, vLLM or uploaded to Hugging Face.
| Unsloth supports | Free Notebooks | Performance | Memory use |
|-----------------|--------------------------------------------------------------------------------------------------------------------------|-------------|----------|
| **Llama-3.1 8b** | [▶️ Start on Colab](https://colab.research.google.com/drive/1Ys44kVvmeZtnICzWz0xgpRnrIOjZAuxp?usp=sharing) | 2.4x faster | 58% less |
| **Phi-3.5 (mini)** | [▶️ Start on Colab](https://colab.research.google.com/drive/1lN6hPQveB_mHSnTOYifygFcrO8C1bxq4?usp=sharing) | 2x faster | 50% less |
| **Gemma-2 9b** | [▶️ Start on Colab](https://colab.research.google.com/drive/1vIrqH5uYDQwsJ4-OO3DErvuv4pBgVwk4?usp=sharing) | 2.4x faster | 58% less |
| **Mistral 7b** | [▶️ Start on Colab](https://colab.research.google.com/drive/1Dyauq4kTZoLewQ1cApceUQVNcnnNTzg_?usp=sharing) | 2.2x faster | 62% less |
| **TinyLlama** | [▶️ Start on Colab](https://colab.research.google.com/drive/1AZghoNBQaMDgWJpi4RbffGM1h6raLUj9?usp=sharing) | 3.9x faster | 74% less |
| **DPO - Zephyr** | [▶️ Start on Colab](https://colab.research.google.com/drive/15vttTpzzVXv_tJwEk-hIcQ0S9FcEWvwP?usp=sharing) | 1.9x faster | 19% less |
- This [conversational notebook](https://colab.research.google.com/drive/1Aau3lgPzeZKQ-98h69CCu1UJcvIBLmy2?usp=sharing) is useful for ShareGPT ChatML / Vicuna templates.
- This [text completion notebook](https://colab.research.google.com/drive/1ef-tab5bhkvWmBOObepl1WgJvfvSzn5Q?usp=sharing) is for raw text. This [DPO notebook](https://colab.research.google.com/drive/15vttTpzzVXv_tJwEk-hIcQ0S9FcEWvwP?usp=sharing) replicates Zephyr.
- \* Kaggle has 2x T4s, but we use 1. Due to overhead, 1x T4 is 5x faster.
# Qwen2.5-14B
## Introduction
Qwen2.5 is the latest series of Qwen large language models. For Qwen2.5, we release a number of base language models and instruction-tuned language models ranging from 0.5 to 72 billion parameters. Qwen2.5 brings the following improvements upon Qwen2:
- Significantly **more knowledge** and has greatly improved capabilities in **coding** and **mathematics**, thanks to our specialized expert models in these domains.
- Significant improvements in **instruction following**, **generating long texts** (over 8K tokens), **understanding structured data** (e.g, tables), and **generating structured outputs** especially JSON. **More resilient to the diversity of system prompts**, enhancing role-play implementation and condition-setting for chatbots.
- **Long-context Support** up to 128K tokens and can generate up to 8K tokens.
- **Multilingual support** for over 29 languages, including Chinese, English, French, Spanish, Portuguese, German, Italian, Russian, Japanese, Korean, Vietnamese, Thai, Arabic, and more.
**This repo contains the base 14B Qwen2.5 model**, which has the following features:
- Type: Causal Language Models
- Training Stage: Pretraining
- Architecture: transformers with RoPE, SwiGLU, RMSNorm, and Attention QKV bias
- Number of Parameters: 14.7B
- Number of Paramaters (Non-Embedding): 13.1B
- Number of Layers: 48
- Number of Attention Heads (GQA): 40 for Q and 8 for KV
- Context Length: 131,072 tokens
**We do not recommend using base language models for conversations.** Instead, you can apply post-training, e.g., SFT, RLHF, continued pretraining, etc., on this model.
For more details, please refer to our [blog](https://qwenlm.github.io/blog/qwen2.5/), [GitHub](https://github.com/QwenLM/Qwen2.5), and [Documentation](https://qwen.readthedocs.io/en/latest/).
## Requirements
The code of Qwen2.5 has been in the latest Hugging face `transformers` and we advise you to use the latest version of `transformers`.
With `transformers<4.37.0`, you will encounter the following error:
```
KeyError: 'qwen2'
```
## Evaluation & Performance
Detailed evaluation results are reported in this [📑 blog](https://qwenlm.github.io/blog/qwen2.5/).
For requirements on GPU memory and the respective throughput, see results [here](https://qwen.readthedocs.io/en/latest/benchmark/speed_benchmark.html).
## Citation
If you find our work helpful, feel free to give us a cite.
```
@misc{qwen2.5,
title = {Qwen2.5: A Party of Foundation Models},
url = {https://qwenlm.github.io/blog/qwen2.5/},
author = {Qwen Team},
month = {September},
year = {2024}
}
@article{qwen2,
title={Qwen2 Technical Report},
author={An Yang and Baosong Yang and Binyuan Hui and Bo Zheng and Bowen Yu and Chang Zhou and Chengpeng Li and Chengyuan Li and Dayiheng Liu and Fei Huang and Guanting Dong and Haoran Wei and Huan Lin and Jialong Tang and Jialin Wang and Jian Yang and Jianhong Tu and Jianwei Zhang and Jianxin Ma and Jin Xu and Jingren Zhou and Jinze Bai and Jinzheng He and Junyang Lin and Kai Dang and Keming Lu and Keqin Chen and Kexin Yang and Mei Li and Mingfeng Xue and Na Ni and Pei Zhang and Peng Wang and Ru Peng and Rui Men and Ruize Gao and Runji Lin and Shijie Wang and Shuai Bai and Sinan Tan and Tianhang Zhu and Tianhao Li and Tianyu Liu and Wenbin Ge and Xiaodong Deng and Xiaohuan Zhou and Xingzhang Ren and Xinyu Zhang and Xipin Wei and Xuancheng Ren and Yang Fan and Yang Yao and Yichang Zhang and Yu Wan and Yunfei Chu and Yuqiong Liu and Zeyu Cui and Zhenru Zhang and Zhihao Fan},
journal={arXiv preprint arXiv:2407.10671},
year={2024}
}
```
|
unsloth/Qwen2.5-3B-bnb-4bit
|
unsloth
| 2025-04-28T04:13:50Z | 1,439 | 2 |
transformers
|
[
"transformers",
"safetensors",
"qwen2",
"text-generation",
"unsloth",
"zho",
"eng",
"fra",
"spa",
"por",
"deu",
"ita",
"rus",
"jpn",
"kor",
"vie",
"tha",
"ara",
"arxiv:2407.10671",
"base_model:Qwen/Qwen2.5-3B",
"base_model:quantized:Qwen/Qwen2.5-3B",
"license:other",
"autotrain_compatible",
"text-generation-inference",
"endpoints_compatible",
"4-bit",
"bitsandbytes",
"region:us"
] |
text-generation
| 2024-09-18T22:16:35Z |
---
base_model: Qwen/Qwen2.5-3B
language:
- zho
- eng
- fra
- spa
- por
- deu
- ita
- rus
- jpn
- kor
- vie
- tha
- ara
library_name: transformers
license: other
tags:
- unsloth
- transformers
---
# Finetune Llama 3.1, Gemma 2, Mistral 2-5x faster with 70% less memory via Unsloth!
We have a Qwen 2.5 (all model sizes) [free Google Colab Tesla T4 notebook](https://colab.research.google.com/drive/1Kose-ucXO1IBaZq5BvbwWieuubP7hxvQ?usp=sharing).
Also a [Qwen 2.5 conversational style notebook](https://colab.research.google.com/drive/1qN1CEalC70EO1wGKhNxs1go1W9So61R5?usp=sharing).
[<img src="https://raw.githubusercontent.com/unslothai/unsloth/main/images/Discord%20button.png" width="200"/>](https://discord.gg/unsloth)
[<img src="https://raw.githubusercontent.com/unslothai/unsloth/main/images/unsloth%20made%20with%20love.png" width="200"/>](https://github.com/unslothai/unsloth)
## ✨ Finetune for Free
All notebooks are **beginner friendly**! Add your dataset, click "Run All", and you'll get a 2x faster finetuned model which can be exported to GGUF, vLLM or uploaded to Hugging Face.
| Unsloth supports | Free Notebooks | Performance | Memory use |
|-----------------|--------------------------------------------------------------------------------------------------------------------------|-------------|----------|
| **Llama-3.1 8b** | [▶️ Start on Colab](https://colab.research.google.com/drive/1Ys44kVvmeZtnICzWz0xgpRnrIOjZAuxp?usp=sharing) | 2.4x faster | 58% less |
| **Phi-3.5 (mini)** | [▶️ Start on Colab](https://colab.research.google.com/drive/1lN6hPQveB_mHSnTOYifygFcrO8C1bxq4?usp=sharing) | 2x faster | 50% less |
| **Gemma-2 9b** | [▶️ Start on Colab](https://colab.research.google.com/drive/1vIrqH5uYDQwsJ4-OO3DErvuv4pBgVwk4?usp=sharing) | 2.4x faster | 58% less |
| **Mistral 7b** | [▶️ Start on Colab](https://colab.research.google.com/drive/1Dyauq4kTZoLewQ1cApceUQVNcnnNTzg_?usp=sharing) | 2.2x faster | 62% less |
| **TinyLlama** | [▶️ Start on Colab](https://colab.research.google.com/drive/1AZghoNBQaMDgWJpi4RbffGM1h6raLUj9?usp=sharing) | 3.9x faster | 74% less |
| **DPO - Zephyr** | [▶️ Start on Colab](https://colab.research.google.com/drive/15vttTpzzVXv_tJwEk-hIcQ0S9FcEWvwP?usp=sharing) | 1.9x faster | 19% less |
- This [conversational notebook](https://colab.research.google.com/drive/1Aau3lgPzeZKQ-98h69CCu1UJcvIBLmy2?usp=sharing) is useful for ShareGPT ChatML / Vicuna templates.
- This [text completion notebook](https://colab.research.google.com/drive/1ef-tab5bhkvWmBOObepl1WgJvfvSzn5Q?usp=sharing) is for raw text. This [DPO notebook](https://colab.research.google.com/drive/15vttTpzzVXv_tJwEk-hIcQ0S9FcEWvwP?usp=sharing) replicates Zephyr.
- \* Kaggle has 2x T4s, but we use 1. Due to overhead, 1x T4 is 5x faster.
# Qwen2.5-3B
## Introduction
Qwen2.5 is the latest series of Qwen large language models. For Qwen2.5, we release a number of base language models and instruction-tuned language models ranging from 0.5 to 72 billion parameters. Qwen2.5 brings the following improvements upon Qwen2:
- Significantly **more knowledge** and has greatly improved capabilities in **coding** and **mathematics**, thanks to our specialized expert models in these domains.
- Significant improvements in **instruction following**, **generating long texts** (over 8K tokens), **understanding structured data** (e.g, tables), and **generating structured outputs** especially JSON. **More resilient to the diversity of system prompts**, enhancing role-play implementation and condition-setting for chatbots.
- **Long-context Support** up to 128K tokens and can generate up to 8K tokens.
- **Multilingual support** for over 29 languages, including Chinese, English, French, Spanish, Portuguese, German, Italian, Russian, Japanese, Korean, Vietnamese, Thai, Arabic, and more.
**This repo contains the base 3B Qwen2.5 model**, which has the following features:
- Type: Causal Language Models
- Training Stage: Pretraining
- Architecture: transformers with RoPE, SwiGLU, RMSNorm, Attention QKV bias and tied word embeddings
- Number of Parameters: 3.09B
- Number of Paramaters (Non-Embedding): 2.77B
- Number of Layers: 36
- Number of Attention Heads (GQA): 16 for Q and 2 for KV
- Context Length: Full 32,768 tokens
**We do not recommend using base language models for conversations.** Instead, you can apply post-training, e.g., SFT, RLHF, continued pretraining, etc., on this model.
For more details, please refer to our [blog](https://qwenlm.github.io/blog/qwen2.5/), [GitHub](https://github.com/QwenLM/Qwen2.5), and [Documentation](https://qwen.readthedocs.io/en/latest/).
## Requirements
The code of Qwen2.5 has been in the latest Hugging face `transformers` and we advise you to use the latest version of `transformers`.
With `transformers<4.37.0`, you will encounter the following error:
```
KeyError: 'qwen2'
```
## Evaluation & Performance
Detailed evaluation results are reported in this [📑 blog](https://qwenlm.github.io/blog/qwen2.5/).
For requirements on GPU memory and the respective throughput, see results [here](https://qwen.readthedocs.io/en/latest/benchmark/speed_benchmark.html).
## Citation
If you find our work helpful, feel free to give us a cite.
```
@misc{qwen2.5,
title = {Qwen2.5: A Party of Foundation Models},
url = {https://qwenlm.github.io/blog/qwen2.5/},
author = {Qwen Team},
month = {September},
year = {2024}
}
@article{qwen2,
title={Qwen2 Technical Report},
author={An Yang and Baosong Yang and Binyuan Hui and Bo Zheng and Bowen Yu and Chang Zhou and Chengpeng Li and Chengyuan Li and Dayiheng Liu and Fei Huang and Guanting Dong and Haoran Wei and Huan Lin and Jialong Tang and Jialin Wang and Jian Yang and Jianhong Tu and Jianwei Zhang and Jianxin Ma and Jin Xu and Jingren Zhou and Jinze Bai and Jinzheng He and Junyang Lin and Kai Dang and Keming Lu and Keqin Chen and Kexin Yang and Mei Li and Mingfeng Xue and Na Ni and Pei Zhang and Peng Wang and Ru Peng and Rui Men and Ruize Gao and Runji Lin and Shijie Wang and Shuai Bai and Sinan Tan and Tianhang Zhu and Tianhao Li and Tianyu Liu and Wenbin Ge and Xiaodong Deng and Xiaohuan Zhou and Xingzhang Ren and Xinyu Zhang and Xipin Wei and Xuancheng Ren and Yang Fan and Yang Yao and Yichang Zhang and Yu Wan and Yunfei Chu and Yuqiong Liu and Zeyu Cui and Zhenru Zhang and Zhihao Fan},
journal={arXiv preprint arXiv:2407.10671},
year={2024}
}
```
|
ranranrunforit/Reinforce-CartPolev1
|
ranranrunforit
| 2025-04-28T04:13:45Z | 0 | 0 | null |
[
"CartPole-v1",
"reinforce",
"reinforcement-learning",
"custom-implementation",
"deep-rl-class",
"model-index",
"region:us"
] |
reinforcement-learning
| 2025-04-28T04:13:35Z |
---
tags:
- CartPole-v1
- reinforce
- reinforcement-learning
- custom-implementation
- deep-rl-class
model-index:
- name: Reinforce-CartPolev1
results:
- task:
type: reinforcement-learning
name: reinforcement-learning
dataset:
name: CartPole-v1
type: CartPole-v1
metrics:
- type: mean_reward
value: 500.00 +/- 0.00
name: mean_reward
verified: false
---
# **Reinforce** Agent playing **CartPole-v1**
This is a trained model of a **Reinforce** agent playing **CartPole-v1** .
To learn to use this model and train yours check Unit 4 of the Deep Reinforcement Learning Course: https://huggingface.co/deep-rl-course/unit4/introduction
|
unsloth/Qwen2.5-0.5B
|
unsloth
| 2025-04-28T04:13:24Z | 6,981 | 9 |
transformers
|
[
"transformers",
"safetensors",
"qwen2",
"text-generation",
"unsloth",
"zho",
"eng",
"fra",
"spa",
"por",
"deu",
"ita",
"rus",
"jpn",
"kor",
"vie",
"tha",
"ara",
"arxiv:2407.10671",
"base_model:Qwen/Qwen2.5-0.5B",
"base_model:finetune:Qwen/Qwen2.5-0.5B",
"license:apache-2.0",
"autotrain_compatible",
"text-generation-inference",
"endpoints_compatible",
"region:us"
] |
text-generation
| 2024-09-18T21:15:04Z |
---
base_model: Qwen/Qwen2.5-0.5B
language:
- zho
- eng
- fra
- spa
- por
- deu
- ita
- rus
- jpn
- kor
- vie
- tha
- ara
library_name: transformers
license: apache-2.0
tags:
- unsloth
- transformers
---
# Finetune Llama 3.1, Gemma 2, Mistral 2-5x faster with 70% less memory via Unsloth!
We have a Qwen 2.5 (all model sizes) [free Google Colab Tesla T4 notebook](https://colab.research.google.com/drive/1Kose-ucXO1IBaZq5BvbwWieuubP7hxvQ?usp=sharing).
Also a [Qwen 2.5 conversational style notebook](https://colab.research.google.com/drive/1qN1CEalC70EO1wGKhNxs1go1W9So61R5?usp=sharing).
[<img src="https://raw.githubusercontent.com/unslothai/unsloth/main/images/Discord%20button.png" width="200"/>](https://discord.gg/unsloth)
[<img src="https://raw.githubusercontent.com/unslothai/unsloth/main/images/unsloth%20made%20with%20love.png" width="200"/>](https://github.com/unslothai/unsloth)
## ✨ Finetune for Free
All notebooks are **beginner friendly**! Add your dataset, click "Run All", and you'll get a 2x faster finetuned model which can be exported to GGUF, vLLM or uploaded to Hugging Face.
| Unsloth supports | Free Notebooks | Performance | Memory use |
|-----------------|--------------------------------------------------------------------------------------------------------------------------|-------------|----------|
| **Llama-3.1 8b** | [▶️ Start on Colab](https://colab.research.google.com/drive/1Ys44kVvmeZtnICzWz0xgpRnrIOjZAuxp?usp=sharing) | 2.4x faster | 58% less |
| **Phi-3.5 (mini)** | [▶️ Start on Colab](https://colab.research.google.com/drive/1lN6hPQveB_mHSnTOYifygFcrO8C1bxq4?usp=sharing) | 2x faster | 50% less |
| **Gemma-2 9b** | [▶️ Start on Colab](https://colab.research.google.com/drive/1vIrqH5uYDQwsJ4-OO3DErvuv4pBgVwk4?usp=sharing) | 2.4x faster | 58% less |
| **Mistral 7b** | [▶️ Start on Colab](https://colab.research.google.com/drive/1Dyauq4kTZoLewQ1cApceUQVNcnnNTzg_?usp=sharing) | 2.2x faster | 62% less |
| **TinyLlama** | [▶️ Start on Colab](https://colab.research.google.com/drive/1AZghoNBQaMDgWJpi4RbffGM1h6raLUj9?usp=sharing) | 3.9x faster | 74% less |
| **DPO - Zephyr** | [▶️ Start on Colab](https://colab.research.google.com/drive/15vttTpzzVXv_tJwEk-hIcQ0S9FcEWvwP?usp=sharing) | 1.9x faster | 19% less |
- This [conversational notebook](https://colab.research.google.com/drive/1Aau3lgPzeZKQ-98h69CCu1UJcvIBLmy2?usp=sharing) is useful for ShareGPT ChatML / Vicuna templates.
- This [text completion notebook](https://colab.research.google.com/drive/1ef-tab5bhkvWmBOObepl1WgJvfvSzn5Q?usp=sharing) is for raw text. This [DPO notebook](https://colab.research.google.com/drive/15vttTpzzVXv_tJwEk-hIcQ0S9FcEWvwP?usp=sharing) replicates Zephyr.
- \* Kaggle has 2x T4s, but we use 1. Due to overhead, 1x T4 is 5x faster.
# Qwen2.5-0.5B
## Introduction
Qwen2.5 is the latest series of Qwen large language models. For Qwen2.5, we release a number of base language models and instruction-tuned language models ranging from 0.5 to 72 billion parameters. Qwen2.5 brings the following improvements upon Qwen2:
- Significantly **more knowledge** and has greatly improved capabilities in **coding** and **mathematics**, thanks to our specialized expert models in these domains.
- Significant improvements in **instruction following**, **generating long texts** (over 8K tokens), **understanding structured data** (e.g, tables), and **generating structured outputs** especially JSON. **More resilient to the diversity of system prompts**, enhancing role-play implementation and condition-setting for chatbots.
- **Long-context Support** up to 128K tokens and can generate up to 8K tokens.
- **Multilingual support** for over 29 languages, including Chinese, English, French, Spanish, Portuguese, German, Italian, Russian, Japanese, Korean, Vietnamese, Thai, Arabic, and more.
**This repo contains the base 0.5B Qwen2.5 model**, which has the following features:
- Type: Causal Language Models
- Training Stage: Pretraining
- Architecture: transformers with RoPE, SwiGLU, RMSNorm, Attention QKV bias and tied word embeddings
- Number of Parameters: 0.49B
- Number of Paramaters (Non-Embedding): 0.36B
- Number of Layers: 24
- Number of Attention Heads (GQA): 14 for Q and 2 for KV
- Context Length: Full 32,768 tokens
**We do not recommend using base language models for conversations.** Instead, you can apply post-training, e.g., SFT, RLHF, continued pretraining, etc., on this model.
For more details, please refer to our [blog](https://qwenlm.github.io/blog/qwen2.5/), [GitHub](https://github.com/QwenLM/Qwen2.5), and [Documentation](https://qwen.readthedocs.io/en/latest/).
## Requirements
The code of Qwen2.5 has been in the latest Hugging face `transformers` and we advise you to use the latest version of `transformers`.
With `transformers<4.37.0`, you will encounter the following error:
```
KeyError: 'qwen2'
```
## Evaluation & Performance
Detailed evaluation results are reported in this [📑 blog](https://qwenlm.github.io/blog/qwen2.5/).
For requirements on GPU memory and the respective throughput, see results [here](https://qwen.readthedocs.io/en/latest/benchmark/speed_benchmark.html).
## Citation
If you find our work helpful, feel free to give us a cite.
```
@misc{qwen2.5,
title = {Qwen2.5: A Party of Foundation Models},
url = {https://qwenlm.github.io/blog/qwen2.5/},
author = {Qwen Team},
month = {September},
year = {2024}
}
@article{qwen2,
title={Qwen2 Technical Report},
author={An Yang and Baosong Yang and Binyuan Hui and Bo Zheng and Bowen Yu and Chang Zhou and Chengpeng Li and Chengyuan Li and Dayiheng Liu and Fei Huang and Guanting Dong and Haoran Wei and Huan Lin and Jialong Tang and Jialin Wang and Jian Yang and Jianhong Tu and Jianwei Zhang and Jianxin Ma and Jin Xu and Jingren Zhou and Jinze Bai and Jinzheng He and Junyang Lin and Kai Dang and Keming Lu and Keqin Chen and Kexin Yang and Mei Li and Mingfeng Xue and Na Ni and Pei Zhang and Peng Wang and Ru Peng and Rui Men and Ruize Gao and Runji Lin and Shijie Wang and Shuai Bai and Sinan Tan and Tianhang Zhu and Tianhao Li and Tianyu Liu and Wenbin Ge and Xiaodong Deng and Xiaohuan Zhou and Xingzhang Ren and Xinyu Zhang and Xipin Wei and Xuancheng Ren and Yang Fan and Yang Yao and Yichang Zhang and Yu Wan and Yunfei Chu and Yuqiong Liu and Zeyu Cui and Zhenru Zhang and Zhihao Fan},
journal={arXiv preprint arXiv:2407.10671},
year={2024}
}
```
|
secmlr/SWE-BENCH-2k-generation-enrich-500-localization-combine_32b-generation-localization-combine
|
secmlr
| 2025-04-28T04:13:19Z | 0 | 0 |
transformers
|
[
"transformers",
"safetensors",
"qwen2",
"text-generation",
"llama-factory",
"full",
"generated_from_trainer",
"conversational",
"base_model:Qwen/Qwen2.5-Coder-32B-Instruct",
"base_model:finetune:Qwen/Qwen2.5-Coder-32B-Instruct",
"license:apache-2.0",
"autotrain_compatible",
"text-generation-inference",
"endpoints_compatible",
"region:us"
] |
text-generation
| 2025-04-27T07:10:34Z |
---
library_name: transformers
license: apache-2.0
base_model: Qwen/Qwen2.5-Coder-32B-Instruct
tags:
- llama-factory
- full
- generated_from_trainer
model-index:
- name: SWE-BENCH-2k-generation-enrich-500-localization-combine_32b-generation-localization-combine
results: []
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# SWE-BENCH-2k-generation-enrich-500-localization-combine_32b-generation-localization-combine
This model is a fine-tuned version of [Qwen/Qwen2.5-Coder-32B-Instruct](https://huggingface.co/Qwen/Qwen2.5-Coder-32B-Instruct) on the SWE-BENCH-2k-generation-enrich-500-localization-combine dataset.
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 1e-05
- train_batch_size: 1
- eval_batch_size: 8
- seed: 42
- distributed_type: multi-GPU
- num_devices: 4
- gradient_accumulation_steps: 12
- total_train_batch_size: 48
- total_eval_batch_size: 32
- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: cosine
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 3.0
### Training results
### Framework versions
- Transformers 4.46.1
- Pytorch 2.5.1+cu124
- Datasets 2.20.0
- Tokenizers 0.20.3
|
Miladawr/Sia
|
Miladawr
| 2025-04-28T04:07:20Z | 0 | 0 | null |
[
"license:cc-by-nc-nd-3.0",
"region:us"
] | null | 2025-04-28T04:07:20Z |
---
license: cc-by-nc-nd-3.0
---
|
fedovtt/92d8ba38-ad64-4656-afdd-f2c3f98c6009
|
fedovtt
| 2025-04-28T04:05:08Z | 0 | 0 |
peft
|
[
"peft",
"safetensors",
"qwen2",
"axolotl",
"generated_from_trainer",
"base_model:Qwen/Qwen2.5-1.5B-Instruct",
"base_model:adapter:Qwen/Qwen2.5-1.5B-Instruct",
"license:apache-2.0",
"8-bit",
"bitsandbytes",
"region:us"
] | null | 2025-04-28T03:58:15Z |
---
library_name: peft
license: apache-2.0
base_model: Qwen/Qwen2.5-1.5B-Instruct
tags:
- axolotl
- generated_from_trainer
model-index:
- name: 92d8ba38-ad64-4656-afdd-f2c3f98c6009
results: []
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
[<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl)
<details><summary>See axolotl config</summary>
axolotl version: `0.4.1`
```yaml
adapter: lora
base_model: Qwen/Qwen2.5-1.5B-Instruct
bf16: true
chat_template: llama3
dataset_prepared_path: null
datasets:
- data_files:
- f2392decb627cf18_train_data.json
ds_type: json
format: custom
path: /workspace/input_data/f2392decb627cf18_train_data.json
type:
field_input: statements
field_instruction: quiz
field_output: solution_text
format: '{instruction} {input}'
no_input_format: '{instruction}'
system_format: '{system}'
system_prompt: ''
debug: null
deepspeed: null
early_stopping_patience: null
eval_max_new_tokens: 128
eval_table_size: null
evals_per_epoch: 1
flash_attention: true
fp16: null
fsdp: null
fsdp_config: null
gradient_accumulation_steps: 1
gradient_checkpointing: true
gradient_clipping: 0.5
group_by_length: false
hub_model_id: fedovtt/92d8ba38-ad64-4656-afdd-f2c3f98c6009
hub_repo: null
hub_strategy: end
hub_token: null
learning_rate: 5.0e-06
load_in_4bit: false
load_in_8bit: true
local_rank: null
logging_steps: 1
lora_alpha: 64
lora_dropout: 0.05
lora_fan_in_fan_out: null
lora_model_dir: null
lora_r: 32
lora_target_linear: true
lr_scheduler: cosine
max_steps: 200
micro_batch_size: 8
mixed_precision: bf16
mlflow_experiment_name: /tmp/f2392decb627cf18_train_data.json
model_type: AutoModelForCausalLM
num_epochs: 1
optimizer: adamw_bnb_8bit
output_dir: miner_id_24
pad_to_sequence_len: true
resume_from_checkpoint: null
s2_attention: null
sample_packing: false
saves_per_epoch: 1
sequence_len: 1024
strict: false
tf32: false
tokenizer_type: AutoTokenizer
train_on_inputs: false
trust_remote_code: true
val_set_size: 0.05
wandb_entity: null
wandb_mode: online
wandb_name: a54f4409-dd56-46d7-8e17-1d233ee1e00a
wandb_project: s56-1
wandb_run: your_name
wandb_runid: a54f4409-dd56-46d7-8e17-1d233ee1e00a
warmup_steps: 5
weight_decay: 0.01
xformers_attention: true
```
</details><br>
# 92d8ba38-ad64-4656-afdd-f2c3f98c6009
This model is a fine-tuned version of [Qwen/Qwen2.5-1.5B-Instruct](https://huggingface.co/Qwen/Qwen2.5-1.5B-Instruct) on the None dataset.
It achieves the following results on the evaluation set:
- Loss: 0.1194
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 5e-06
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- optimizer: Use OptimizerNames.ADAMW_BNB with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: cosine
- lr_scheduler_warmup_steps: 5
- training_steps: 200
### Training results
| Training Loss | Epoch | Step | Validation Loss |
|:-------------:|:------:|:----:|:---------------:|
| 0.1156 | 0.0468 | 200 | 0.1194 |
### Framework versions
- PEFT 0.13.2
- Transformers 4.46.0
- Pytorch 2.5.0+cu124
- Datasets 3.0.1
- Tokenizers 0.20.1
|
OldKingMeister/Qwen2.5-1.5B-Instruct-YaRN
|
OldKingMeister
| 2025-04-28T04:01:29Z | 29 | 1 |
transformers
|
[
"transformers",
"safetensors",
"qwen2",
"text-generation",
"chat",
"conversational",
"zho",
"eng",
"fra",
"spa",
"por",
"deu",
"ita",
"rus",
"jpn",
"kor",
"vie",
"tha",
"ara",
"arxiv:2407.10671",
"base_model:Qwen/Qwen2.5-1.5B",
"base_model:finetune:Qwen/Qwen2.5-1.5B",
"license:apache-2.0",
"autotrain_compatible",
"text-generation-inference",
"endpoints_compatible",
"region:us"
] |
text-generation
| 2025-04-04T09:23:38Z |
---
license: apache-2.0
license_link: https://huggingface.co/Qwen/Qwen2.5-1.5B-Instruct/blob/main/LICENSE
language:
- zho
- eng
- fra
- spa
- por
- deu
- ita
- rus
- jpn
- kor
- vie
- tha
- ara
pipeline_tag: text-generation
base_model: Qwen/Qwen2.5-1.5B
tags:
- chat
library_name: transformers
---
# Modification
Context length extension technique ***YaRN*** is enabled for this model as illustrated in the [original development doc](https://qwen.readthedocs.io/en/latest/deployment/vllm.html#extended-context-support).
Before is the original model card.
# Qwen2.5-1.5B-Instruct
## Introduction
Qwen2.5 is the latest series of Qwen large language models. For Qwen2.5, we release a number of base language models and instruction-tuned language models ranging from 0.5 to 72 billion parameters. Qwen2.5 brings the following improvements upon Qwen2:
- Significantly **more knowledge** and has greatly improved capabilities in **coding** and **mathematics**, thanks to our specialized expert models in these domains.
- Significant improvements in **instruction following**, **generating long texts** (over 8K tokens), **understanding structured data** (e.g, tables), and **generating structured outputs** especially JSON. **More resilient to the diversity of system prompts**, enhancing role-play implementation and condition-setting for chatbots.
- **Long-context Support** up to 128K tokens and can generate up to 8K tokens.
- **Multilingual support** for over 29 languages, including Chinese, English, French, Spanish, Portuguese, German, Italian, Russian, Japanese, Korean, Vietnamese, Thai, Arabic, and more.
**This repo contains the instruction-tuned 1.5B Qwen2.5 model**, which has the following features:
- Type: Causal Language Models
- Training Stage: Pretraining & Post-training
- Architecture: transformers with RoPE, SwiGLU, RMSNorm, Attention QKV bias and tied word embeddings
- Number of Parameters: 1.54B
- Number of Paramaters (Non-Embedding): 1.31B
- Number of Layers: 28
- Number of Attention Heads (GQA): 12 for Q and 2 for KV
- Context Length: Full 32,768 tokens and generation 8192 tokens
For more details, please refer to our [blog](https://qwenlm.github.io/blog/qwen2.5/), [GitHub](https://github.com/QwenLM/Qwen2.5), and [Documentation](https://qwen.readthedocs.io/en/latest/).
## Requirements
The code of Qwen2.5 has been in the latest Hugging face `transformers` and we advise you to use the latest version of `transformers`.
With `transformers<4.37.0`, you will encounter the following error:
```
KeyError: 'qwen2'
```
## Quickstart
Here provides a code snippet with `apply_chat_template` to show you how to load the tokenizer and model and how to generate contents.
```python
from transformers import AutoModelForCausalLM, AutoTokenizer
model_name = "Qwen/Qwen2.5-1.5B-Instruct"
model = AutoModelForCausalLM.from_pretrained(
model_name,
torch_dtype="auto",
device_map="auto"
)
tokenizer = AutoTokenizer.from_pretrained(model_name)
prompt = "Give me a short introduction to large language model."
messages = [
{"role": "system", "content": "You are Qwen, created by Alibaba Cloud. You are a helpful assistant."},
{"role": "user", "content": prompt}
]
text = tokenizer.apply_chat_template(
messages,
tokenize=False,
add_generation_prompt=True
)
model_inputs = tokenizer([text], return_tensors="pt").to(model.device)
generated_ids = model.generate(
**model_inputs,
max_new_tokens=512
)
generated_ids = [
output_ids[len(input_ids):] for input_ids, output_ids in zip(model_inputs.input_ids, generated_ids)
]
response = tokenizer.batch_decode(generated_ids, skip_special_tokens=True)[0]
```
## Evaluation & Performance
Detailed evaluation results are reported in this [📑 blog](https://qwenlm.github.io/blog/qwen2.5/).
For requirements on GPU memory and the respective throughput, see results [here](https://qwen.readthedocs.io/en/latest/benchmark/speed_benchmark.html).
## Citation
If you find our work helpful, feel free to give us a cite.
```
@misc{qwen2.5,
title = {Qwen2.5: A Party of Foundation Models},
url = {https://qwenlm.github.io/blog/qwen2.5/},
author = {Qwen Team},
month = {September},
year = {2024}
}
@article{qwen2,
title={Qwen2 Technical Report},
author={An Yang and Baosong Yang and Binyuan Hui and Bo Zheng and Bowen Yu and Chang Zhou and Chengpeng Li and Chengyuan Li and Dayiheng Liu and Fei Huang and Guanting Dong and Haoran Wei and Huan Lin and Jialong Tang and Jialin Wang and Jian Yang and Jianhong Tu and Jianwei Zhang and Jianxin Ma and Jin Xu and Jingren Zhou and Jinze Bai and Jinzheng He and Junyang Lin and Kai Dang and Keming Lu and Keqin Chen and Kexin Yang and Mei Li and Mingfeng Xue and Na Ni and Pei Zhang and Peng Wang and Ru Peng and Rui Men and Ruize Gao and Runji Lin and Shijie Wang and Shuai Bai and Sinan Tan and Tianhang Zhu and Tianhao Li and Tianyu Liu and Wenbin Ge and Xiaodong Deng and Xiaohuan Zhou and Xingzhang Ren and Xinyu Zhang and Xipin Wei and Xuancheng Ren and Yang Fan and Yang Yao and Yichang Zhang and Yu Wan and Yunfei Chu and Yuqiong Liu and Zeyu Cui and Zhenru Zhang and Zhihao Fan},
journal={arXiv preprint arXiv:2407.10671},
year={2024}
}
```
|
marialvsantiago/ab63f62d-53ae-4662-869d-5349dfc10eca
|
marialvsantiago
| 2025-04-28T04:00:30Z | 0 | 0 |
peft
|
[
"peft",
"safetensors",
"qwen2",
"axolotl",
"generated_from_trainer",
"base_model:Qwen/Qwen2.5-1.5B-Instruct",
"base_model:adapter:Qwen/Qwen2.5-1.5B-Instruct",
"license:apache-2.0",
"4-bit",
"bitsandbytes",
"region:us"
] | null | 2025-04-28T03:58:02Z |
---
library_name: peft
license: apache-2.0
base_model: Qwen/Qwen2.5-1.5B-Instruct
tags:
- axolotl
- generated_from_trainer
model-index:
- name: ab63f62d-53ae-4662-869d-5349dfc10eca
results: []
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
[<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl)
<details><summary>See axolotl config</summary>
axolotl version: `0.4.1`
```yaml
adapter: lora
base_model: Qwen/Qwen2.5-1.5B-Instruct
bf16: true
chat_template: llama3
dataset_prepared_path: null
datasets:
- data_files:
- f2392decb627cf18_train_data.json
ds_type: json
format: custom
path: /workspace/input_data/f2392decb627cf18_train_data.json
type:
field_input: statements
field_instruction: quiz
field_output: solution_text
format: '{instruction} {input}'
no_input_format: '{instruction}'
system_format: '{system}'
system_prompt: ''
debug: null
deepspeed: null
early_stopping_patience: null
eval_max_new_tokens: 128
eval_table_size: null
evals_per_epoch: 1
flash_attention: true
fp16: null
fsdp: null
fsdp_config: null
gradient_accumulation_steps: 1
gradient_checkpointing: true
gradient_clipping: 0.5
group_by_length: false
hub_model_id: marialvsantiago/ab63f62d-53ae-4662-869d-5349dfc10eca
hub_repo: null
hub_strategy: end
hub_token: null
learning_rate: 5.0e-06
load_in_4bit: true
load_in_8bit: false
local_rank: null
logging_steps: 1
lora_alpha: 64
lora_dropout: 0.05
lora_fan_in_fan_out: null
lora_model_dir: null
lora_r: 32
lora_target_linear: true
lr_scheduler: cosine
max_steps: 200
micro_batch_size: 8
mixed_precision: bf16
mlflow_experiment_name: /tmp/f2392decb627cf18_train_data.json
model_type: AutoModelForCausalLM
num_epochs: 1
optimizer: adamw_bnb_8bit
output_dir: miner_id_24
pad_to_sequence_len: true
resume_from_checkpoint: null
s2_attention: null
sample_packing: false
saves_per_epoch: 1
sequence_len: 1024
strict: false
tf32: false
tokenizer_type: AutoTokenizer
train_on_inputs: false
trust_remote_code: true
val_set_size: 0.05
wandb_entity: null
wandb_mode: online
wandb_name: a54f4409-dd56-46d7-8e17-1d233ee1e00a
wandb_project: s56-33
wandb_run: your_name
wandb_runid: a54f4409-dd56-46d7-8e17-1d233ee1e00a
warmup_steps: 5
weight_decay: 0.01
xformers_attention: true
```
</details><br>
# ab63f62d-53ae-4662-869d-5349dfc10eca
This model is a fine-tuned version of [Qwen/Qwen2.5-1.5B-Instruct](https://huggingface.co/Qwen/Qwen2.5-1.5B-Instruct) on the None dataset.
It achieves the following results on the evaluation set:
- Loss: 0.1221
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 5e-06
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- optimizer: Use OptimizerNames.ADAMW_BNB with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: cosine
- lr_scheduler_warmup_steps: 5
- training_steps: 200
### Training results
| Training Loss | Epoch | Step | Validation Loss |
|:-------------:|:------:|:----:|:---------------:|
| 0.124 | 0.0468 | 200 | 0.1221 |
### Framework versions
- PEFT 0.13.2
- Transformers 4.46.0
- Pytorch 2.5.0+cu124
- Datasets 3.0.1
- Tokenizers 0.20.1
|
TianZhou621/bert-finetuned-ner
|
TianZhou621
| 2025-04-28T04:00:25Z | 0 | 0 |
transformers
|
[
"transformers",
"tensorboard",
"safetensors",
"bert",
"token-classification",
"generated_from_trainer",
"dataset:conll2003",
"base_model:google-bert/bert-base-cased",
"base_model:finetune:google-bert/bert-base-cased",
"license:apache-2.0",
"model-index",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] |
token-classification
| 2025-04-28T03:48:18Z |
---
library_name: transformers
license: apache-2.0
base_model: bert-base-cased
tags:
- generated_from_trainer
datasets:
- conll2003
metrics:
- precision
- recall
- f1
- accuracy
model-index:
- name: bert-finetuned-ner
results:
- task:
name: Token Classification
type: token-classification
dataset:
name: conll2003
type: conll2003
config: conll2003
split: validation
args: conll2003
metrics:
- name: Precision
type: precision
value: 0.9468494066521812
- name: Recall
type: recall
value: 0.9533826994278021
- name: F1
type: f1
value: 0.9501048218029349
- name: Accuracy
type: accuracy
value: 0.9916280518671391
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# bert-finetuned-ner
This model is a fine-tuned version of [bert-base-cased](https://huggingface.co/bert-base-cased) on the conll2003 dataset.
It achieves the following results on the evaluation set:
- Loss: 0.0401
- Precision: 0.9468
- Recall: 0.9534
- F1: 0.9501
- Accuracy: 0.9916
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 2e-05
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- num_epochs: 3
### Training results
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
|:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|:--------:|
| 0.0527 | 1.0 | 1756 | 0.0465 | 0.9239 | 0.9352 | 0.9295 | 0.9883 |
| 0.0242 | 2.0 | 3512 | 0.0418 | 0.9449 | 0.9473 | 0.9461 | 0.9911 |
| 0.0123 | 3.0 | 5268 | 0.0401 | 0.9468 | 0.9534 | 0.9501 | 0.9916 |
### Framework versions
- Transformers 4.51.3
- Pytorch 2.6.0+cu124
- Datasets 3.5.0
- Tokenizers 0.21.1
|
DDpsrETgya3rbD/hshuu7t
|
DDpsrETgya3rbD
| 2025-04-28T03:54:02Z | 0 | 0 | null |
[
"license:apache-2.0",
"region:us"
] | null | 2025-04-28T03:54:02Z |
---
license: apache-2.0
---
|
ranranrunforit/dqn-SpaceInvadersNoFrameskip-v4
|
ranranrunforit
| 2025-04-28T03:50:16Z | 0 | 0 |
stable-baselines3
|
[
"stable-baselines3",
"SpaceInvadersNoFrameskip-v4",
"deep-reinforcement-learning",
"reinforcement-learning",
"model-index",
"region:us"
] |
reinforcement-learning
| 2025-04-28T03:49:49Z |
---
library_name: stable-baselines3
tags:
- SpaceInvadersNoFrameskip-v4
- deep-reinforcement-learning
- reinforcement-learning
- stable-baselines3
model-index:
- name: DQN
results:
- task:
type: reinforcement-learning
name: reinforcement-learning
dataset:
name: SpaceInvadersNoFrameskip-v4
type: SpaceInvadersNoFrameskip-v4
metrics:
- type: mean_reward
value: 497.50 +/- 110.16
name: mean_reward
verified: false
---
# **DQN** Agent playing **SpaceInvadersNoFrameskip-v4**
This is a trained model of a **DQN** agent playing **SpaceInvadersNoFrameskip-v4**
using the [stable-baselines3 library](https://github.com/DLR-RM/stable-baselines3)
and the [RL Zoo](https://github.com/DLR-RM/rl-baselines3-zoo).
The RL Zoo is a training framework for Stable Baselines3
reinforcement learning agents,
with hyperparameter optimization and pre-trained agents included.
## Usage (with SB3 RL Zoo)
RL Zoo: https://github.com/DLR-RM/rl-baselines3-zoo<br/>
SB3: https://github.com/DLR-RM/stable-baselines3<br/>
SB3 Contrib: https://github.com/Stable-Baselines-Team/stable-baselines3-contrib
SBX (SB3 + Jax): https://github.com/araffin/sbx
Install the RL Zoo (with SB3 and SB3-Contrib):
```bash
pip install rl_zoo3
```
```
# Download model and save it into the logs/ folder
python -m rl_zoo3.load_from_hub --algo dqn --env SpaceInvadersNoFrameskip-v4 -orga ranranrunforit -f logs/
python -m rl_zoo3.enjoy --algo dqn --env SpaceInvadersNoFrameskip-v4 -f logs/
```
If you installed the RL Zoo3 via pip (`pip install rl_zoo3`), from anywhere you can do:
```
python -m rl_zoo3.load_from_hub --algo dqn --env SpaceInvadersNoFrameskip-v4 -orga ranranrunforit -f logs/
python -m rl_zoo3.enjoy --algo dqn --env SpaceInvadersNoFrameskip-v4 -f logs/
```
## Training (with the RL Zoo)
```
python -m rl_zoo3.train --algo dqn --env SpaceInvadersNoFrameskip-v4 -f logs/
# Upload the model and generate video (when possible)
python -m rl_zoo3.push_to_hub --algo dqn --env SpaceInvadersNoFrameskip-v4 -f logs/ -orga ranranrunforit
```
## Hyperparameters
```python
OrderedDict([('batch_size', 32),
('buffer_size', 100000),
('env_wrapper',
['stable_baselines3.common.atari_wrappers.AtariWrapper']),
('exploration_final_eps', 0.01),
('exploration_fraction', 0.1),
('frame_stack', 4),
('gradient_steps', 1),
('learning_rate', 0.0001),
('learning_starts', 100000),
('n_timesteps', 1000000.0),
('optimize_memory_usage', False),
('policy', 'CnnPolicy'),
('target_update_interval', 1000),
('train_freq', 4),
('normalize', False)])
```
# Environment Arguments
```python
{'render_mode': 'rgb_array'}
```
|
Frowning/Test-I-24B
|
Frowning
| 2025-04-28T03:47:36Z | 0 | 0 |
transformers
|
[
"transformers",
"safetensors",
"mistral",
"text-generation",
"mergekit",
"merge",
"conversational",
"arxiv:2408.07990",
"base_model:Gryphe/Pantheon-RP-1.8-24b-Small-3.1",
"base_model:merge:Gryphe/Pantheon-RP-1.8-24b-Small-3.1",
"base_model:TroyDoesAI/BlackSheep-24B",
"base_model:merge:TroyDoesAI/BlackSheep-24B",
"base_model:aixonlab/Eurydice-24b-v2",
"base_model:merge:aixonlab/Eurydice-24b-v2",
"autotrain_compatible",
"text-generation-inference",
"endpoints_compatible",
"region:us"
] |
text-generation
| 2025-04-28T03:37:05Z |
---
base_model:
- TroyDoesAI/BlackSheep-24B
- Gryphe/Pantheon-RP-1.8-24b-Small-3.1
- aixonlab/Eurydice-24b-v2
library_name: transformers
tags:
- mergekit
- merge
---
# merge
This is a merge of pre-trained language models created using [mergekit](https://github.com/cg123/mergekit).
## Merge Details
### Merge Method
This model was merged using the [SCE](https://arxiv.org/abs/2408.07990) merge method using [TroyDoesAI/BlackSheep-24B](https://huggingface.co/TroyDoesAI/BlackSheep-24B) as a base.
### Models Merged
The following models were included in the merge:
* [Gryphe/Pantheon-RP-1.8-24b-Small-3.1](https://huggingface.co/Gryphe/Pantheon-RP-1.8-24b-Small-3.1)
* [aixonlab/Eurydice-24b-v2](https://huggingface.co/aixonlab/Eurydice-24b-v2)
### Configuration
The following YAML configuration was used to produce this model:
```yaml
base_model: TroyDoesAI/BlackSheep-24B
models:
- model: Gryphe/Pantheon-RP-1.8-24b-Small-3.1
parameters:
weight: 0.45
- model: aixonlab/Eurydice-24b-v2
parameters:
weight: 0.55
merge_method: sce
dtype: bfloat16
parameters:
normalize: true
select_topk: 0.8
tokenizer:
source: union
```
|
Kojar3D/Ray
|
Kojar3D
| 2025-04-28T03:45:48Z | 0 | 0 | null |
[
"license:apache-2.0",
"region:us"
] | null | 2025-04-28T03:45:48Z |
---
license: apache-2.0
---
|
Yany9538/medical-question-model
|
Yany9538
| 2025-04-28T03:43:48Z | 0 | 0 |
transformers
|
[
"transformers",
"safetensors",
"bert",
"text-classification",
"arxiv:1910.09700",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] |
text-classification
| 2025-04-28T03:43:14Z |
---
library_name: transformers
tags: []
---
# Model Card for Model ID
<!-- Provide a quick summary of what the model is/does. -->
## Model Details
### Model Description
<!-- Provide a longer summary of what this model is. -->
This is the model card of a 🤗 transformers model that has been pushed on the Hub. This model card has been automatically generated.
- **Developed by:** [More Information Needed]
- **Funded by [optional]:** [More Information Needed]
- **Shared by [optional]:** [More Information Needed]
- **Model type:** [More Information Needed]
- **Language(s) (NLP):** [More Information Needed]
- **License:** [More Information Needed]
- **Finetuned from model [optional]:** [More Information Needed]
### Model Sources [optional]
<!-- Provide the basic links for the model. -->
- **Repository:** [More Information Needed]
- **Paper [optional]:** [More Information Needed]
- **Demo [optional]:** [More Information Needed]
## Uses
<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
### Direct Use
<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
[More Information Needed]
### Downstream Use [optional]
<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
[More Information Needed]
### Out-of-Scope Use
<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
[More Information Needed]
## Bias, Risks, and Limitations
<!-- This section is meant to convey both technical and sociotechnical limitations. -->
[More Information Needed]
### Recommendations
<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
## How to Get Started with the Model
Use the code below to get started with the model.
[More Information Needed]
## Training Details
### Training Data
<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
[More Information Needed]
### Training Procedure
<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
#### Preprocessing [optional]
[More Information Needed]
#### Training Hyperparameters
- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
#### Speeds, Sizes, Times [optional]
<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
[More Information Needed]
## Evaluation
<!-- This section describes the evaluation protocols and provides the results. -->
### Testing Data, Factors & Metrics
#### Testing Data
<!-- This should link to a Dataset Card if possible. -->
[More Information Needed]
#### Factors
<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
[More Information Needed]
#### Metrics
<!-- These are the evaluation metrics being used, ideally with a description of why. -->
[More Information Needed]
### Results
[More Information Needed]
#### Summary
## Model Examination [optional]
<!-- Relevant interpretability work for the model goes here -->
[More Information Needed]
## Environmental Impact
<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
- **Hardware Type:** [More Information Needed]
- **Hours used:** [More Information Needed]
- **Cloud Provider:** [More Information Needed]
- **Compute Region:** [More Information Needed]
- **Carbon Emitted:** [More Information Needed]
## Technical Specifications [optional]
### Model Architecture and Objective
[More Information Needed]
### Compute Infrastructure
[More Information Needed]
#### Hardware
[More Information Needed]
#### Software
[More Information Needed]
## Citation [optional]
<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
**BibTeX:**
[More Information Needed]
**APA:**
[More Information Needed]
## Glossary [optional]
<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
[More Information Needed]
## More Information [optional]
[More Information Needed]
## Model Card Authors [optional]
[More Information Needed]
## Model Card Contact
[More Information Needed]
|
rippertnt/HyperCLOVAX-SEED-Text-Instruct-1.5B-Q4_K_M-GGUF
|
rippertnt
| 2025-04-28T03:43:00Z | 29 | 0 | null |
[
"gguf",
"llama",
"llama-cpp",
"gguf-my-repo",
"base_model:naver-hyperclovax/HyperCLOVAX-SEED-Text-Instruct-1.5B",
"base_model:quantized:naver-hyperclovax/HyperCLOVAX-SEED-Text-Instruct-1.5B",
"license:other",
"endpoints_compatible",
"region:us",
"conversational"
] | null | 2025-04-24T04:03:35Z |
---
base_model: naver-hyperclovax/HyperCLOVAX-SEED-Text-Instruct-1.5B
license: other
license_name: hyperclovax-seed
license_link: LICENSE
tags:
- llama-cpp
- gguf-my-repo
---
# rippertnt/HyperCLOVAX-SEED-Text-Instruct-1.5B-Q4_K_M-GGUF
This model was converted to GGUF format from [`naver-hyperclovax/HyperCLOVAX-SEED-Text-Instruct-1.5B`](https://huggingface.co/naver-hyperclovax/HyperCLOVAX-SEED-Text-Instruct-1.5B) using llama.cpp via the ggml.ai's [GGUF-my-repo](https://huggingface.co/spaces/ggml-org/gguf-my-repo) space.
Refer to the [original model card](https://huggingface.co/naver-hyperclovax/HyperCLOVAX-SEED-Text-Instruct-1.5B) for more details on the model.
## Use with llama.cpp
Install llama.cpp through brew (works on Mac and Linux)
```bash
brew install llama.cpp
```
Invoke the llama.cpp server or the CLI.
### CLI:
```bash
llama-cli --hf-repo rippertnt/HyperCLOVAX-SEED-Text-Instruct-1.5B-Q4_K_M-GGUF --hf-file hyperclovax-seed-text-instruct-1.5b-q4_k_m.gguf -p "The meaning to life and the universe is"
```
### Server:
```bash
llama-server --hf-repo rippertnt/HyperCLOVAX-SEED-Text-Instruct-1.5B-Q4_K_M-GGUF --hf-file hyperclovax-seed-text-instruct-1.5b-q4_k_m.gguf -c 2048
```
Note: You can also use this checkpoint directly through the [usage steps](https://github.com/ggerganov/llama.cpp?tab=readme-ov-file#usage) listed in the Llama.cpp repo as well.
Step 1: Clone llama.cpp from GitHub.
```
git clone https://github.com/ggerganov/llama.cpp
```
Step 2: Move into the llama.cpp folder and build it with `LLAMA_CURL=1` flag along with other hardware-specific flags (for ex: LLAMA_CUDA=1 for Nvidia GPUs on Linux).
```
cd llama.cpp && LLAMA_CURL=1 make
```
Step 3: Run inference through the main binary.
```
./llama-cli --hf-repo rippertnt/HyperCLOVAX-SEED-Text-Instruct-1.5B-Q4_K_M-GGUF --hf-file hyperclovax-seed-text-instruct-1.5b-q4_k_m.gguf -p "The meaning to life and the universe is"
```
or
```
./llama-server --hf-repo rippertnt/HyperCLOVAX-SEED-Text-Instruct-1.5B-Q4_K_M-GGUF --hf-file hyperclovax-seed-text-instruct-1.5b-q4_k_m.gguf -c 2048
```
|
JohnConnor123/Meta-Llama-3-8B-BNB-4bit
|
JohnConnor123
| 2025-04-28T03:39:46Z | 0 | 0 | null |
[
"safetensors",
"llama",
"en",
"base_model:meta-llama/Meta-Llama-3-8B",
"base_model:quantized:meta-llama/Meta-Llama-3-8B",
"4-bit",
"bitsandbytes",
"region:us"
] | null | 2025-04-28T03:34:42Z |
---
language: en
base_model: meta-llama/Meta-Llama-3-8B
---
> ## **This quantization was done using the [quantization-benchmark](https://github.com/JohnConnor123/quantization-benchmark) framework**
## Model Details
Meta developed and released the Meta Llama 3 family of large language models (LLMs), a collection of pretrained and instruction tuned generative text models in 8 and 70B sizes. The Llama 3 instruction tuned models are optimized for dialogue use cases and outperform many of the available open source chat models on common industry benchmarks. Further, in developing these models, we took great care to optimize helpfulness and safety.
**Model developers** Meta
**Variations** Llama 3 comes in two sizes — 8B and 70B parameters — in pre-trained and instruction tuned variants.
**Input** Models input text only.
**Output** Models generate text and code only.
**Model Architecture** Llama 3 is an auto-regressive language model that uses an optimized transformer architecture. The tuned versions use supervised fine-tuning (SFT) and reinforcement learning with human feedback (RLHF) to align with human preferences for helpfulness and safety.
<table>
<tr>
<td>
</td>
<td><strong>Training Data</strong>
</td>
<td><strong>Params</strong>
</td>
<td><strong>Context length</strong>
</td>
<td><strong>GQA</strong>
</td>
<td><strong>Token count</strong>
</td>
<td><strong>Knowledge cutoff</strong>
</td>
</tr>
<tr>
<td rowspan="2" >Llama 3
</td>
<td rowspan="2" >A new mix of publicly available online data.
</td>
<td>8B
</td>
<td>8k
</td>
<td>Yes
</td>
<td rowspan="2" >15T+
</td>
<td>March, 2023
</td>
</tr>
<tr>
<td>70B
</td>
<td>8k
</td>
<td>Yes
</td>
<td>December, 2023
</td>
</tr>
</table>
**Llama 3 family of models**. Token counts refer to pretraining data only. Both the 8 and 70B versions use Grouped-Query Attention (GQA) for improved inference scalability.
**Model Release Date** April 18, 2024.
**Status** This is a static model trained on an offline dataset. Future versions of the tuned models will be released as we improve model safety with community feedback.
**License** A custom commercial license is available at: [https://llama.meta.com/llama3/license](https://llama.meta.com/llama3/license)
Where to send questions or comments about the model Instructions on how to provide feedback or comments on the model can be found in the model [README](https://github.com/meta-llama/llama3). For more technical information about generation parameters and recipes for how to use Llama 3 in applications, please go [here](https://github.com/meta-llama/llama-recipes).
## Intended Use
**Intended Use Cases** Llama 3 is intended for commercial and research use in English. Instruction tuned models are intended for assistant-like chat, whereas pretrained models can be adapted for a variety of natural language generation tasks.
**Out-of-scope** Use in any manner that violates applicable laws or regulations (including trade compliance laws). Use in any other way that is prohibited by the Acceptable Use Policy and Llama 3 Community License. Use in languages other than English**.
**Note: Developers may fine-tune Llama 3 models for languages beyond English provided they comply with the Llama 3 Community License and the Acceptable Use Policy.
## How to use
This repository contains two versions of Meta-Llama-3-8B, for use with transformers and with the original `llama3` codebase.
### Use with transformers
See the snippet below for usage with Transformers:
```python
>>> import transformers
>>> import torch
>>> model_id = "meta-llama/Meta-Llama-3-8B"
>>> pipeline = transformers.pipeline(
"text-generation", model=model_id, model_kwargs={"torch_dtype": torch.bfloat16}, device_map="auto"
)
>>> pipeline("Hey how are you doing today?")
```
### Use with `llama3`
Please, follow the instructions in the [repository](https://github.com/meta-llama/llama3).
To download Original checkpoints, see the example command below leveraging `huggingface-cli`:
```
huggingface-cli download meta-llama/Meta-Llama-3-8B --include "original/*" --local-dir Meta-Llama-3-8B
```
For Hugging Face support, we recommend using transformers or TGI, but a similar command works.
## Hardware and Software
**Training Factors** We used custom training libraries, Meta's Research SuperCluster, and production clusters for pretraining. Fine-tuning, annotation, and evaluation were also performed on third-party cloud compute.
**Carbon Footprint Pretraining utilized a cumulative** 7.7M GPU hours of computation on hardware of type H100-80GB (TDP of 700W). Estimated total emissions were 2290 tCO2eq, 100% of which were offset by Meta’s sustainability program.
<table>
<tr>
<td>
</td>
<td><strong>Time (GPU hours)</strong>
</td>
<td><strong>Power Consumption (W)</strong>
</td>
<td><strong>Carbon Emitted(tCO2eq)</strong>
</td>
</tr>
<tr>
<td>Llama 3 8B
</td>
<td>1.3M
</td>
<td>700
</td>
<td>390
</td>
</tr>
<tr>
<td>Llama 3 70B
</td>
<td>6.4M
</td>
<td>700
</td>
<td>1900
</td>
</tr>
<tr>
<td>Total
</td>
<td>7.7M
</td>
<td>
</td>
<td>2290
</td>
</tr>
</table>
**CO2 emissions during pre-training**. Time: total GPU time required for training each model. Power Consumption: peak power capacity per GPU device for the GPUs used adjusted for power usage efficiency. 100% of the emissions are directly offset by Meta's sustainability program, and because we are openly releasing these models, the pretraining costs do not need to be incurred by others.
## Training Data
**Overview** Llama 3 was pretrained on over 15 trillion tokens of data from publicly available sources. The fine-tuning data includes publicly available instruction datasets, as well as over 10M human-annotated examples. Neither the pretraining nor the fine-tuning datasets include Meta user data.
**Data Freshness** The pretraining data has a cutoff of March 2023 for the 8B and December 2023 for the 70B models respectively.
## Benchmarks
In this section, we report the results for Llama 3 models on standard automatic benchmarks. For all the evaluations, we use our internal evaluations library. For details on the methodology see [here](https://github.com/meta-llama/llama3/blob/main/eval_methodology.md).
### Base pretrained models
<table>
<tr>
<td><strong>Category</strong>
</td>
<td><strong>Benchmark</strong>
</td>
<td><strong>Llama 3 8B</strong>
</td>
<td><strong>Llama2 7B</strong>
</td>
<td><strong>Llama2 13B</strong>
</td>
<td><strong>Llama 3 70B</strong>
</td>
<td><strong>Llama2 70B</strong>
</td>
</tr>
<tr>
<td rowspan="6" >General
</td>
<td>MMLU (5-shot)
</td>
<td>66.6
</td>
<td>45.7
</td>
<td>53.8
</td>
<td>79.5
</td>
<td>69.7
</td>
</tr>
<tr>
<td>AGIEval English (3-5 shot)
</td>
<td>45.9
</td>
<td>28.8
</td>
<td>38.7
</td>
<td>63.0
</td>
<td>54.8
</td>
</tr>
<tr>
<td>CommonSenseQA (7-shot)
</td>
<td>72.6
</td>
<td>57.6
</td>
<td>67.6
</td>
<td>83.8
</td>
<td>78.7
</td>
</tr>
<tr>
<td>Winogrande (5-shot)
</td>
<td>76.1
</td>
<td>73.3
</td>
<td>75.4
</td>
<td>83.1
</td>
<td>81.8
</td>
</tr>
<tr>
<td>BIG-Bench Hard (3-shot, CoT)
</td>
<td>61.1
</td>
<td>38.1
</td>
<td>47.0
</td>
<td>81.3
</td>
<td>65.7
</td>
</tr>
<tr>
<td>ARC-Challenge (25-shot)
</td>
<td>78.6
</td>
<td>53.7
</td>
<td>67.6
</td>
<td>93.0
</td>
<td>85.3
</td>
</tr>
<tr>
<td>Knowledge reasoning
</td>
<td>TriviaQA-Wiki (5-shot)
</td>
<td>78.5
</td>
<td>72.1
</td>
<td>79.6
</td>
<td>89.7
</td>
<td>87.5
</td>
</tr>
<tr>
<td rowspan="4" >Reading comprehension
</td>
<td>SQuAD (1-shot)
</td>
<td>76.4
</td>
<td>72.2
</td>
<td>72.1
</td>
<td>85.6
</td>
<td>82.6
</td>
</tr>
<tr>
<td>QuAC (1-shot, F1)
</td>
<td>44.4
</td>
<td>39.6
</td>
<td>44.9
</td>
<td>51.1
</td>
<td>49.4
</td>
</tr>
<tr>
<td>BoolQ (0-shot)
</td>
<td>75.7
</td>
<td>65.5
</td>
<td>66.9
</td>
<td>79.0
</td>
<td>73.1
</td>
</tr>
<tr>
<td>DROP (3-shot, F1)
</td>
<td>58.4
</td>
<td>37.9
</td>
<td>49.8
</td>
<td>79.7
</td>
<td>70.2
</td>
</tr>
</table>
### Instruction tuned models
<table>
<tr>
<td><strong>Benchmark</strong>
</td>
<td><strong>Llama 3 8B</strong>
</td>
<td><strong>Llama 2 7B</strong>
</td>
<td><strong>Llama 2 13B</strong>
</td>
<td><strong>Llama 3 70B</strong>
</td>
<td><strong>Llama 2 70B</strong>
</td>
</tr>
<tr>
<td>MMLU (5-shot)
</td>
<td>68.4
</td>
<td>34.1
</td>
<td>47.8
</td>
<td>82.0
</td>
<td>52.9
</td>
</tr>
<tr>
<td>GPQA (0-shot)
</td>
<td>34.2
</td>
<td>21.7
</td>
<td>22.3
</td>
<td>39.5
</td>
<td>21.0
</td>
</tr>
<tr>
<td>HumanEval (0-shot)
</td>
<td>62.2
</td>
<td>7.9
</td>
<td>14.0
</td>
<td>81.7
</td>
<td>25.6
</td>
</tr>
<tr>
<td>GSM-8K (8-shot, CoT)
</td>
<td>79.6
</td>
<td>25.7
</td>
<td>77.4
</td>
<td>93.0
</td>
<td>57.5
</td>
</tr>
<tr>
<td>MATH (4-shot, CoT)
</td>
<td>30.0
</td>
<td>3.8
</td>
<td>6.7
</td>
<td>50.4
</td>
<td>11.6
</td>
</tr>
</table>
### Responsibility & Safety
We believe that an open approach to AI leads to better, safer products, faster innovation, and a bigger overall market. We are committed to Responsible AI development and took a series of steps to limit misuse and harm and support the open source community.
Foundation models are widely capable technologies that are built to be used for a diverse range of applications. They are not designed to meet every developer preference on safety levels for all use cases, out-of-the-box, as those by their nature will differ across different applications.
Rather, responsible LLM-application deployment is achieved by implementing a series of safety best practices throughout the development of such applications, from the model pre-training, fine-tuning and the deployment of systems composed of safeguards to tailor the safety needs specifically to the use case and audience.
As part of the Llama 3 release, we updated our [Responsible Use Guide](https://llama.meta.com/responsible-use-guide/) to outline the steps and best practices for developers to implement model and system level safety for their application. We also provide a set of resources including [Meta Llama Guard 2](https://llama.meta.com/purple-llama/) and [Code Shield](https://llama.meta.com/purple-llama/) safeguards. These tools have proven to drastically reduce residual risks of LLM Systems, while maintaining a high level of helpfulness. We encourage developers to tune and deploy these safeguards according to their needs and we provide a [reference implementation](https://github.com/meta-llama/llama-recipes/tree/main/recipes/responsible_ai) to get you started.
#### Llama 3-Instruct
As outlined in the Responsible Use Guide, some trade-off between model helpfulness and model alignment is likely unavoidable. Developers should exercise discretion about how to weigh the benefits of alignment and helpfulness for their specific use case and audience. Developers should be mindful of residual risks when using Llama models and leverage additional safety tools as needed to reach the right safety bar for their use case.
<span style="text-decoration:underline;">Safety</span>
For our instruction tuned model, we conducted extensive red teaming exercises, performed adversarial evaluations and implemented safety mitigations techniques to lower residual risks. As with any Large Language Model, residual risks will likely remain and we recommend that developers assess these risks in the context of their use case. In parallel, we are working with the community to make AI safety benchmark standards transparent, rigorous and interpretable.
<span style="text-decoration:underline;">Refusals</span>
In addition to residual risks, we put a great emphasis on model refusals to benign prompts. Over-refusing not only can impact the user experience but could even be harmful in certain contexts as well. We’ve heard the feedback from the developer community and improved our fine tuning to ensure that Llama 3 is significantly less likely to falsely refuse to answer prompts than Llama 2.
We built internal benchmarks and developed mitigations to limit false refusals making Llama 3 our most helpful model to date.
#### Responsible release
In addition to responsible use considerations outlined above, we followed a rigorous process that requires us to take extra measures against misuse and critical risks before we make our release decision.
Misuse
If you access or use Llama 3, you agree to the Acceptable Use Policy. The most recent copy of this policy can be found at [https://llama.meta.com/llama3/use-policy/](https://llama.meta.com/llama3/use-policy/).
#### Critical risks
<span style="text-decoration:underline;">CBRNE</span> (Chemical, Biological, Radiological, Nuclear, and high yield Explosives)
We have conducted a two fold assessment of the safety of the model in this area:
* Iterative testing during model training to assess the safety of responses related to CBRNE threats and other adversarial risks.
* Involving external CBRNE experts to conduct an uplift test assessing the ability of the model to accurately provide expert knowledge and reduce barriers to potential CBRNE misuse, by reference to what can be achieved using web search (without the model).
### <span style="text-decoration:underline;">Cyber Security </span>
We have evaluated Llama 3 with CyberSecEval, Meta’s cybersecurity safety eval suite, measuring Llama 3’s propensity to suggest insecure code when used as a coding assistant, and Llama 3’s propensity to comply with requests to help carry out cyber attacks, where attacks are defined by the industry standard MITRE ATT&CK cyber attack ontology. On our insecure coding and cyber attacker helpfulness tests, Llama 3 behaved in the same range or safer than models of [equivalent coding capability](https://huggingface.co/spaces/facebook/CyberSecEval).
### <span style="text-decoration:underline;">Child Safety</span>
Child Safety risk assessments were conducted using a team of experts, to assess the model’s capability to produce outputs that could result in Child Safety risks and inform on any necessary and appropriate risk mitigations via fine tuning. We leveraged those expert red teaming sessions to expand the coverage of our evaluation benchmarks through Llama 3 model development. For Llama 3, we conducted new in-depth sessions using objective based methodologies to assess the model risks along multiple attack vectors. We also partnered with content specialists to perform red teaming exercises assessing potentially violating content while taking account of market specific nuances or experiences.
### Community
Generative AI safety requires expertise and tooling, and we believe in the strength of the open community to accelerate its progress. We are active members of open consortiums, including the AI Alliance, Partnership in AI and MLCommons, actively contributing to safety standardization and transparency. We encourage the community to adopt taxonomies like the MLCommons Proof of Concept evaluation to facilitate collaboration and transparency on safety and content evaluations. Our Purple Llama tools are open sourced for the community to use and widely distributed across ecosystem partners including cloud service providers. We encourage community contributions to our [Github repository](https://github.com/meta-llama/PurpleLlama).
Finally, we put in place a set of resources including an [output reporting mechanism](https://developers.facebook.com/llama_output_feedback) and [bug bounty program](https://www.facebook.com/whitehat) to continuously improve the Llama technology with the help of the community.
## Ethical Considerations and Limitations
The core values of Llama 3 are openness, inclusivity and helpfulness. It is meant to serve everyone, and to work for a wide range of use cases. It is thus designed to be accessible to people across many different backgrounds, experiences and perspectives. Llama 3 addresses users and their needs as they are, without insertion unnecessary judgment or normativity, while reflecting the understanding that even content that may appear problematic in some cases can serve valuable purposes in others. It respects the dignity and autonomy of all users, especially in terms of the values of free thought and expression that power innovation and progress.
But Llama 3 is a new technology, and like any new technology, there are risks associated with its use. Testing conducted to date has been in English, and has not covered, nor could it cover, all scenarios. For these reasons, as with all LLMs, Llama 3’s potential outputs cannot be predicted in advance, and the model may in some instances produce inaccurate, biased or other objectionable responses to user prompts. Therefore, before deploying any applications of Llama 3 models, developers should perform safety testing and tuning tailored to their specific applications of the model. As outlined in the Responsible Use Guide, we recommend incorporating [Purple Llama](https://github.com/facebookresearch/PurpleLlama) solutions into your workflows and specifically [Llama Guard](https://ai.meta.com/research/publications/llama-guard-llm-based-input-output-safeguard-for-human-ai-conversations/) which provides a base model to filter input and output prompts to layer system-level safety on top of model-level safety.
Please see the Responsible Use Guide available at [http://llama.meta.com/responsible-use-guide](http://llama.meta.com/responsible-use-guide)
## Citation instructions
@article{llama3modelcard,
title={Llama 3 Model Card},
author={AI@Meta},
year={2024},
url = {https://github.com/meta-llama/llama3/blob/main/MODEL_CARD.md}
}
## Contributors
Aaditya Singh; Aaron Grattafiori; Abhimanyu Dubey; Abhinav Jauhri; Abhinav Pandey; Abhishek Kadian; Adam Kelsey; Adi Gangidi; Ahmad Al-Dahle; Ahuva Goldstand; Aiesha Letman; Ajay Menon; Akhil Mathur; Alan Schelten; Alex Vaughan; Amy Yang; Andrei Lupu; Andres Alvarado; Andrew Gallagher; Andrew Gu; Andrew Ho; Andrew Poulton; Andrew Ryan; Angela Fan; Ankit Ramchandani; Anthony Hartshorn; Archi Mitra; Archie Sravankumar; Artem Korenev; Arun Rao; Ashley Gabriel; Ashwin Bharambe; Assaf Eisenman; Aston Zhang; Aurelien Rodriguez; Austen Gregerson; Ava Spataru; Baptiste Roziere; Ben Maurer; Benjamin Leonhardi; Bernie Huang; Bhargavi Paranjape; Bing Liu; Binh Tang; Bobbie Chern; Brani Stojkovic; Brian Fuller; Catalina Mejia Arenas; Chao Zhou; Charlotte Caucheteux; Chaya Nayak; Ching-Hsiang Chu; Chloe Bi; Chris Cai; Chris Cox; Chris Marra; Chris McConnell; Christian Keller; Christoph Feichtenhofer; Christophe Touret; Chunyang Wu; Corinne Wong; Cristian Canton Ferrer; Damien Allonsius; Daniel Kreymer; Daniel Haziza; Daniel Li; Danielle Pintz; Danny Livshits; Danny Wyatt; David Adkins; David Esiobu; David Xu; Davide Testuggine; Delia David; Devi Parikh; Dhruv Choudhary; Dhruv Mahajan; Diana Liskovich; Diego Garcia-Olano; Diego Perino; Dieuwke Hupkes; Dingkang Wang; Dustin Holland; Egor Lakomkin; Elina Lobanova; Xiaoqing Ellen Tan; Emily Dinan; Eric Smith; Erik Brinkman; Esteban Arcaute; Filip Radenovic; Firat Ozgenel; Francesco Caggioni; Frank Seide; Frank Zhang; Gabriel Synnaeve; Gabriella Schwarz; Gabrielle Lee; Gada Badeer; Georgia Anderson; Graeme Nail; Gregoire Mialon; Guan Pang; Guillem Cucurell; Hailey Nguyen; Hannah Korevaar; Hannah Wang; Haroun Habeeb; Harrison Rudolph; Henry Aspegren; Hu Xu; Hugo Touvron; Iga Kozlowska; Igor Molybog; Igor Tufanov; Iliyan Zarov; Imanol Arrieta Ibarra; Irina-Elena Veliche; Isabel Kloumann; Ishan Misra; Ivan Evtimov; Jacob Xu; Jade Copet; Jake Weissman; Jan Geffert; Jana Vranes; Japhet Asher; Jason Park; Jay Mahadeokar; Jean-Baptiste Gaya; Jeet Shah; Jelmer van der Linde; Jennifer Chan; Jenny Hong; Jenya Lee; Jeremy Fu; Jeremy Teboul; Jianfeng Chi; Jianyu Huang; Jie Wang; Jiecao Yu; Joanna Bitton; Joe Spisak; Joelle Pineau; Jon Carvill; Jongsoo Park; Joseph Rocca; Joshua Johnstun; Junteng Jia; Kalyan Vasuden Alwala; Kam Hou U; Kate Plawiak; Kartikeya Upasani; Kaushik Veeraraghavan; Ke Li; Kenneth Heafield; Kevin Stone; Khalid El-Arini; Krithika Iyer; Kshitiz Malik; Kuenley Chiu; Kunal Bhalla; Kyle Huang; Lakshya Garg; Lauren Rantala-Yeary; Laurens van der Maaten; Lawrence Chen; Leandro Silva; Lee Bell; Lei Zhang; Liang Tan; Louis Martin; Lovish Madaan; Luca Wehrstedt; Lukas Blecher; Luke de Oliveira; Madeline Muzzi; Madian Khabsa; Manav Avlani; Mannat Singh; Manohar Paluri; Mark Zuckerberg; Marcin Kardas; Martynas Mankus; Mathew Oldham; Mathieu Rita; Matthew Lennie; Maya Pavlova; Meghan Keneally; Melanie Kambadur; Mihir Patel; Mikayel Samvelyan; Mike Clark; Mike Lewis; Min Si; Mitesh Kumar Singh; Mo Metanat; Mona Hassan; Naman Goyal; Narjes Torabi; Nicolas Usunier; Nikolay Bashlykov; Nikolay Bogoychev; Niladri Chatterji; Ning Dong; Oliver Aobo Yang; Olivier Duchenne; Onur Celebi; Parth Parekh; Patrick Alrassy; Paul Saab; Pavan Balaji; Pedro Rittner; Pengchuan Zhang; Pengwei Li; Petar Vasic; Peter Weng; Polina Zvyagina; Prajjwal Bhargava; Pratik Dubal; Praveen Krishnan; Punit Singh Koura; Qing He; Rachel Rodriguez; Ragavan Srinivasan; Rahul Mitra; Ramon Calderer; Raymond Li; Robert Stojnic; Roberta Raileanu; Robin Battey; Rocky Wang; Rohit Girdhar; Rohit Patel; Romain Sauvestre; Ronnie Polidoro; Roshan Sumbaly; Ross Taylor; Ruan Silva; Rui Hou; Rui Wang; Russ Howes; Ruty Rinott; Saghar Hosseini; Sai Jayesh Bondu; Samyak Datta; Sanjay Singh; Sara Chugh; Sargun Dhillon; Satadru Pan; Sean Bell; Sergey Edunov; Shaoliang Nie; Sharan Narang; Sharath Raparthy; Shaun Lindsay; Sheng Feng; Sheng Shen; Shenghao Lin; Shiva Shankar; Shruti Bhosale; Shun Zhang; Simon Vandenhende; Sinong Wang; Seohyun Sonia Kim; Soumya Batra; Sten Sootla; Steve Kehoe; Suchin Gururangan; Sumit Gupta; Sunny Virk; Sydney Borodinsky; Tamar Glaser; Tamar Herman; Tamara Best; Tara Fowler; Thomas Georgiou; Thomas Scialom; Tianhe Li; Todor Mihaylov; Tong Xiao; Ujjwal Karn; Vedanuj Goswami; Vibhor Gupta; Vignesh Ramanathan; Viktor Kerkez; Vinay Satish Kumar; Vincent Gonguet; Vish Vogeti; Vlad Poenaru; Vlad Tiberiu Mihailescu; Vladan Petrovic; Vladimir Ivanov; Wei Li; Weiwei Chu; Wenhan Xiong; Wenyin Fu; Wes Bouaziz; Whitney Meers; Will Constable; Xavier Martinet; Xiaojian Wu; Xinbo Gao; Xinfeng Xie; Xuchao Jia; Yaelle Goldschlag; Yann LeCun; Yashesh Gaur; Yasmine Babaei; Ye Qi; Yenda Li; Yi Wen; Yiwen Song; Youngjin Nam; Yuchen Hao; Yuchen Zhang; Yun Wang; Yuning Mao; Yuzi He; Zacharie Delpierre Coudert; Zachary DeVito; Zahra Hankir; Zhaoduo Wen; Zheng Yan; Zhengxing Chen; Zhenyu Yang; Zoe Papakipos
## Bitsandbytes quantization config
>{'load_in_4bit': True}
|
infogeo/25194737-fee0-456d-b0eb-0253f6d2e668
|
infogeo
| 2025-04-28T03:37:43Z | 0 | 0 |
peft
|
[
"peft",
"safetensors",
"llama",
"axolotl",
"generated_from_trainer",
"base_model:tokyotech-llm/Llama-3-Swallow-8B-v0.1",
"base_model:adapter:tokyotech-llm/Llama-3-Swallow-8B-v0.1",
"license:llama3",
"4-bit",
"bitsandbytes",
"region:us"
] | null | 2025-04-28T03:31:41Z |
---
library_name: peft
license: llama3
base_model: tokyotech-llm/Llama-3-Swallow-8B-v0.1
tags:
- axolotl
- generated_from_trainer
model-index:
- name: 25194737-fee0-456d-b0eb-0253f6d2e668
results: []
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
[<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl)
<details><summary>See axolotl config</summary>
axolotl version: `0.4.1`
```yaml
absolute_data_files: false
adapter: lora
base_model: tokyotech-llm/Llama-3-Swallow-8B-v0.1
bf16: true
chat_template: llama3
dataset_prepared_path: /workspace/axolotl
datasets:
- data_files:
- 1967260052b4d18f_train_data.json
ds_type: json
format: custom
path: /workspace/input_data/1967260052b4d18f_train_data.json
type:
field_input: system_prompt
field_instruction: problem
field_output: solution
format: '{instruction} {input}'
no_input_format: '{instruction}'
system_format: '{system}'
system_prompt: ''
debug: null
deepspeed: null
early_stopping_patience: null
eval_max_new_tokens: 128
eval_table_size: null
evals_per_epoch: 1
flash_attention: true
fp16: null
fsdp: null
fsdp_config: null
gradient_accumulation_steps: 1
gradient_checkpointing: true
gradient_clipping: 0.55
group_by_length: false
hub_model_id: infogeo/25194737-fee0-456d-b0eb-0253f6d2e668
hub_repo: null
hub_strategy: end
hub_token: null
learning_rate: 1.0e-06
load_in_4bit: true
load_in_8bit: false
local_rank: null
logging_steps: 1
lora_alpha: 64
lora_dropout: 0.05
lora_fan_in_fan_out: null
lora_model_dir: null
lora_r: 32
lora_target_linear: true
lr_scheduler: cosine
max_steps: 150
micro_batch_size: 8
mixed_precision: bf16
mlflow_experiment_name: /tmp/1967260052b4d18f_train_data.json
model_type: AutoModelForCausalLM
num_epochs: 1
optimizer: adamw_bnb_8bit
output_dir: miner_id_24
pad_to_sequence_len: true
resume_from_checkpoint: null
s2_attention: null
sample_packing: false
saves_per_epoch: 1
sequence_len: 1024
special_tokens:
pad_token: <|end_of_text|>
strict: false
tf32: false
tokenizer_type: AutoTokenizer
train_on_inputs: false
trust_remote_code: true
val_set_size: 0.05
wandb_entity: null
wandb_mode: online
wandb_name: 24e0af35-36d6-42f2-84cc-bb9b98b57426
wandb_project: s56-28
wandb_run: your_name
wandb_runid: 24e0af35-36d6-42f2-84cc-bb9b98b57426
warmup_steps: 5
weight_decay: 0.01
xformers_attention: true
```
</details><br>
# 25194737-fee0-456d-b0eb-0253f6d2e668
This model is a fine-tuned version of [tokyotech-llm/Llama-3-Swallow-8B-v0.1](https://huggingface.co/tokyotech-llm/Llama-3-Swallow-8B-v0.1) on the None dataset.
It achieves the following results on the evaluation set:
- Loss: 0.9242
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 1e-06
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- optimizer: Use OptimizerNames.ADAMW_BNB with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: cosine
- lr_scheduler_warmup_steps: 5
- training_steps: 150
### Training results
| Training Loss | Epoch | Step | Validation Loss |
|:-------------:|:------:|:----:|:---------------:|
| 0.8855 | 0.0253 | 150 | 0.9242 |
### Framework versions
- PEFT 0.13.2
- Transformers 4.46.0
- Pytorch 2.5.0+cu124
- Datasets 3.0.1
- Tokenizers 0.20.1
|
mjpsm/participation-image-classifier
|
mjpsm
| 2025-04-28T03:34:07Z | 0 | 0 |
transformers
|
[
"transformers",
"safetensors",
"vit",
"image-classification",
"arxiv:1910.09700",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] |
image-classification
| 2025-04-28T03:32:36Z |
---
library_name: transformers
tags: []
---
# Model Card for Model ID
<!-- Provide a quick summary of what the model is/does. -->
## Model Details
### Model Description
<!-- Provide a longer summary of what this model is. -->
This is the model card of a 🤗 transformers model that has been pushed on the Hub. This model card has been automatically generated.
- **Developed by:** [More Information Needed]
- **Funded by [optional]:** [More Information Needed]
- **Shared by [optional]:** [More Information Needed]
- **Model type:** [More Information Needed]
- **Language(s) (NLP):** [More Information Needed]
- **License:** [More Information Needed]
- **Finetuned from model [optional]:** [More Information Needed]
### Model Sources [optional]
<!-- Provide the basic links for the model. -->
- **Repository:** [More Information Needed]
- **Paper [optional]:** [More Information Needed]
- **Demo [optional]:** [More Information Needed]
## Uses
<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
### Direct Use
<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
[More Information Needed]
### Downstream Use [optional]
<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
[More Information Needed]
### Out-of-Scope Use
<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
[More Information Needed]
## Bias, Risks, and Limitations
<!-- This section is meant to convey both technical and sociotechnical limitations. -->
[More Information Needed]
### Recommendations
<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
## How to Get Started with the Model
Use the code below to get started with the model.
[More Information Needed]
## Training Details
### Training Data
<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
[More Information Needed]
### Training Procedure
<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
#### Preprocessing [optional]
[More Information Needed]
#### Training Hyperparameters
- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
#### Speeds, Sizes, Times [optional]
<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
[More Information Needed]
## Evaluation
<!-- This section describes the evaluation protocols and provides the results. -->
### Testing Data, Factors & Metrics
#### Testing Data
<!-- This should link to a Dataset Card if possible. -->
[More Information Needed]
#### Factors
<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
[More Information Needed]
#### Metrics
<!-- These are the evaluation metrics being used, ideally with a description of why. -->
[More Information Needed]
### Results
[More Information Needed]
#### Summary
## Model Examination [optional]
<!-- Relevant interpretability work for the model goes here -->
[More Information Needed]
## Environmental Impact
<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
- **Hardware Type:** [More Information Needed]
- **Hours used:** [More Information Needed]
- **Cloud Provider:** [More Information Needed]
- **Compute Region:** [More Information Needed]
- **Carbon Emitted:** [More Information Needed]
## Technical Specifications [optional]
### Model Architecture and Objective
[More Information Needed]
### Compute Infrastructure
[More Information Needed]
#### Hardware
[More Information Needed]
#### Software
[More Information Needed]
## Citation [optional]
<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
**BibTeX:**
[More Information Needed]
**APA:**
[More Information Needed]
## Glossary [optional]
<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
[More Information Needed]
## More Information [optional]
[More Information Needed]
## Model Card Authors [optional]
[More Information Needed]
## Model Card Contact
[More Information Needed]
|
mradermacher/Qwen2.5-Ultimate-14B-Instruct-i1-GGUF
|
mradermacher
| 2025-04-28T03:33:23Z | 88 | 1 |
transformers
|
[
"transformers",
"gguf",
"mergekit",
"merge",
"zho",
"eng",
"fra",
"spa",
"por",
"deu",
"ita",
"rus",
"jpn",
"kor",
"vie",
"tha",
"ara",
"base_model:qingy2024/Qwen2.5-Ultimate-14B-Instruct",
"base_model:quantized:qingy2024/Qwen2.5-Ultimate-14B-Instruct",
"endpoints_compatible",
"region:us",
"imatrix",
"conversational"
] | null | 2024-12-04T00:24:01Z |
---
base_model: qingy2024/Qwen2.5-Ultimate-14B-Instruct
language:
- zho
- eng
- fra
- spa
- por
- deu
- ita
- rus
- jpn
- kor
- vie
- tha
- ara
library_name: transformers
quantized_by: mradermacher
tags:
- mergekit
- merge
---
## About
<!-- ### quantize_version: 2 -->
<!-- ### output_tensor_quantised: 1 -->
<!-- ### convert_type: hf -->
<!-- ### vocab_type: -->
<!-- ### tags: nicoboss -->
weighted/imatrix quants of https://huggingface.co/qingy2024/Qwen2.5-Ultimate-14B-Instruct
<!-- provided-files -->
static quants are available at https://huggingface.co/mradermacher/Qwen2.5-Ultimate-14B-Instruct-GGUF
## Usage
If you are unsure how to use GGUF files, refer to one of [TheBloke's
READMEs](https://huggingface.co/TheBloke/KafkaLM-70B-German-V0.1-GGUF) for
more details, including on how to concatenate multi-part files.
## Provided Quants
(sorted by size, not necessarily quality. IQ-quants are often preferable over similar sized non-IQ quants)
| Link | Type | Size/GB | Notes |
|:-----|:-----|--------:|:------|
| [GGUF](https://huggingface.co/mradermacher/Qwen2.5-Ultimate-14B-Instruct-i1-GGUF/resolve/main/Qwen2.5-Ultimate-14B-Instruct.i1-IQ1_S.gguf) | i1-IQ1_S | 3.7 | for the desperate |
| [GGUF](https://huggingface.co/mradermacher/Qwen2.5-Ultimate-14B-Instruct-i1-GGUF/resolve/main/Qwen2.5-Ultimate-14B-Instruct.i1-IQ1_M.gguf) | i1-IQ1_M | 4.0 | mostly desperate |
| [GGUF](https://huggingface.co/mradermacher/Qwen2.5-Ultimate-14B-Instruct-i1-GGUF/resolve/main/Qwen2.5-Ultimate-14B-Instruct.i1-IQ2_XXS.gguf) | i1-IQ2_XXS | 4.4 | |
| [GGUF](https://huggingface.co/mradermacher/Qwen2.5-Ultimate-14B-Instruct-i1-GGUF/resolve/main/Qwen2.5-Ultimate-14B-Instruct.i1-IQ2_XS.gguf) | i1-IQ2_XS | 4.8 | |
| [GGUF](https://huggingface.co/mradermacher/Qwen2.5-Ultimate-14B-Instruct-i1-GGUF/resolve/main/Qwen2.5-Ultimate-14B-Instruct.i1-IQ2_S.gguf) | i1-IQ2_S | 5.1 | |
| [GGUF](https://huggingface.co/mradermacher/Qwen2.5-Ultimate-14B-Instruct-i1-GGUF/resolve/main/Qwen2.5-Ultimate-14B-Instruct.i1-IQ2_M.gguf) | i1-IQ2_M | 5.5 | |
| [GGUF](https://huggingface.co/mradermacher/Qwen2.5-Ultimate-14B-Instruct-i1-GGUF/resolve/main/Qwen2.5-Ultimate-14B-Instruct.i1-Q2_K_S.gguf) | i1-Q2_K_S | 5.5 | very low quality |
| [GGUF](https://huggingface.co/mradermacher/Qwen2.5-Ultimate-14B-Instruct-i1-GGUF/resolve/main/Qwen2.5-Ultimate-14B-Instruct.i1-Q2_K.gguf) | i1-Q2_K | 5.9 | IQ3_XXS probably better |
| [GGUF](https://huggingface.co/mradermacher/Qwen2.5-Ultimate-14B-Instruct-i1-GGUF/resolve/main/Qwen2.5-Ultimate-14B-Instruct.i1-IQ3_XXS.gguf) | i1-IQ3_XXS | 6.0 | lower quality |
| [GGUF](https://huggingface.co/mradermacher/Qwen2.5-Ultimate-14B-Instruct-i1-GGUF/resolve/main/Qwen2.5-Ultimate-14B-Instruct.i1-IQ3_XS.gguf) | i1-IQ3_XS | 6.5 | |
| [GGUF](https://huggingface.co/mradermacher/Qwen2.5-Ultimate-14B-Instruct-i1-GGUF/resolve/main/Qwen2.5-Ultimate-14B-Instruct.i1-Q3_K_S.gguf) | i1-Q3_K_S | 6.8 | IQ3_XS probably better |
| [GGUF](https://huggingface.co/mradermacher/Qwen2.5-Ultimate-14B-Instruct-i1-GGUF/resolve/main/Qwen2.5-Ultimate-14B-Instruct.i1-IQ3_S.gguf) | i1-IQ3_S | 6.8 | beats Q3_K* |
| [GGUF](https://huggingface.co/mradermacher/Qwen2.5-Ultimate-14B-Instruct-i1-GGUF/resolve/main/Qwen2.5-Ultimate-14B-Instruct.i1-IQ3_M.gguf) | i1-IQ3_M | 7.0 | |
| [GGUF](https://huggingface.co/mradermacher/Qwen2.5-Ultimate-14B-Instruct-i1-GGUF/resolve/main/Qwen2.5-Ultimate-14B-Instruct.i1-Q3_K_M.gguf) | i1-Q3_K_M | 7.4 | IQ3_S probably better |
| [GGUF](https://huggingface.co/mradermacher/Qwen2.5-Ultimate-14B-Instruct-i1-GGUF/resolve/main/Qwen2.5-Ultimate-14B-Instruct.i1-Q3_K_L.gguf) | i1-Q3_K_L | 8.0 | IQ3_M probably better |
| [GGUF](https://huggingface.co/mradermacher/Qwen2.5-Ultimate-14B-Instruct-i1-GGUF/resolve/main/Qwen2.5-Ultimate-14B-Instruct.i1-IQ4_XS.gguf) | i1-IQ4_XS | 8.2 | |
| [GGUF](https://huggingface.co/mradermacher/Qwen2.5-Ultimate-14B-Instruct-i1-GGUF/resolve/main/Qwen2.5-Ultimate-14B-Instruct.i1-Q4_0_4_4.gguf) | i1-Q4_0_4_4 | 8.6 | fast on arm, low quality |
| [GGUF](https://huggingface.co/mradermacher/Qwen2.5-Ultimate-14B-Instruct-i1-GGUF/resolve/main/Qwen2.5-Ultimate-14B-Instruct.i1-Q4_0_4_8.gguf) | i1-Q4_0_4_8 | 8.6 | fast on arm+i8mm, low quality |
| [GGUF](https://huggingface.co/mradermacher/Qwen2.5-Ultimate-14B-Instruct-i1-GGUF/resolve/main/Qwen2.5-Ultimate-14B-Instruct.i1-Q4_0_8_8.gguf) | i1-Q4_0_8_8 | 8.6 | fast on arm+sve, low quality |
| [GGUF](https://huggingface.co/mradermacher/Qwen2.5-Ultimate-14B-Instruct-i1-GGUF/resolve/main/Qwen2.5-Ultimate-14B-Instruct.i1-Q4_0.gguf) | i1-Q4_0 | 8.6 | fast, low quality |
| [GGUF](https://huggingface.co/mradermacher/Qwen2.5-Ultimate-14B-Instruct-i1-GGUF/resolve/main/Qwen2.5-Ultimate-14B-Instruct.i1-Q4_K_S.gguf) | i1-Q4_K_S | 8.7 | optimal size/speed/quality |
| [GGUF](https://huggingface.co/mradermacher/Qwen2.5-Ultimate-14B-Instruct-i1-GGUF/resolve/main/Qwen2.5-Ultimate-14B-Instruct.i1-Q4_K_M.gguf) | i1-Q4_K_M | 9.1 | fast, recommended |
| [GGUF](https://huggingface.co/mradermacher/Qwen2.5-Ultimate-14B-Instruct-i1-GGUF/resolve/main/Qwen2.5-Ultimate-14B-Instruct.i1-Q5_K_S.gguf) | i1-Q5_K_S | 10.4 | |
| [GGUF](https://huggingface.co/mradermacher/Qwen2.5-Ultimate-14B-Instruct-i1-GGUF/resolve/main/Qwen2.5-Ultimate-14B-Instruct.i1-Q5_K_M.gguf) | i1-Q5_K_M | 10.6 | |
| [GGUF](https://huggingface.co/mradermacher/Qwen2.5-Ultimate-14B-Instruct-i1-GGUF/resolve/main/Qwen2.5-Ultimate-14B-Instruct.i1-Q6_K.gguf) | i1-Q6_K | 12.2 | practically like static Q6_K |
Here is a handy graph by ikawrakow comparing some lower-quality quant
types (lower is better):
And here are Artefact2's thoughts on the matter:
https://gist.github.com/Artefact2/b5f810600771265fc1e39442288e8ec9
## FAQ / Model Request
See https://huggingface.co/mradermacher/model_requests for some answers to
questions you might have and/or if you want some other model quantized.
## Thanks
I thank my company, [nethype GmbH](https://www.nethype.de/), for letting
me use its servers and providing upgrades to my workstation to enable
this work in my free time. Additional thanks to [@nicoboss](https://huggingface.co/nicoboss) for giving me access to his private supercomputer, enabling me to provide many more imatrix quants, at much higher quality, than I would otherwise be able to.
<!-- end -->
|
Jonjew/CatherineDeneuve
|
Jonjew
| 2025-04-28T03:32:29Z | 0 | 0 |
diffusers
|
[
"diffusers",
"text-to-image",
"lora",
"template:diffusion-lora",
"base_model:black-forest-labs/FLUX.1-dev",
"base_model:adapter:black-forest-labs/FLUX.1-dev",
"license:unknown",
"region:us"
] |
text-to-image
| 2025-04-28T03:32:24Z |
---
tags:
- text-to-image
- lora
- diffusers
- template:diffusion-lora
widget:
- text: catherinedeneuve
output:
url: images/1129-catherinedeneuve-Fluxflux1-dev-fp8-50720747.png
base_model: black-forest-labs/FLUX.1-dev
instance_prompt: catherinedeneuve
license: unknown
---
# Catherine Deneuve by cbrescia
<Gallery />
## Model description
FROM https://civitai.com/models/1502212/catherine-deneuve
Please support the creator by donating BUZZ to the creator and LIKING at the page above
Trigger catherinedeneuve
## Trigger words
You should use `catherinedeneuve` to trigger the image generation.
## Download model
Weights for this model are available in Safetensors format.
[Download](/Jonjew/CatherineDeneuve/tree/main) them in the Files & versions tab.
|
elixir4020/my-awesome-model
|
elixir4020
| 2025-04-28T03:29:35Z | 0 | 0 | null |
[
"safetensors",
"pytorch_model_hub_mixin",
"model_hub_mixin",
"region:us"
] | null | 2025-04-28T03:29:08Z |
---
tags:
- pytorch_model_hub_mixin
- model_hub_mixin
---
This model has been pushed to the Hub using the [PytorchModelHubMixin](https://huggingface.co/docs/huggingface_hub/package_reference/mixins#huggingface_hub.PyTorchModelHubMixin) integration:
- Library: [More Information Needed]
- Docs: [More Information Needed]
|
mradermacher/Fusion-14B-Instruct-GGUF
|
mradermacher
| 2025-04-28T03:27:05Z | 27 | 1 |
transformers
|
[
"transformers",
"gguf",
"mergekit",
"merge",
"zho",
"eng",
"fra",
"spa",
"por",
"deu",
"ita",
"rus",
"jpn",
"kor",
"vie",
"tha",
"ara",
"base_model:qingy2024/Fusion-14B-Instruct",
"base_model:quantized:qingy2024/Fusion-14B-Instruct",
"endpoints_compatible",
"region:us",
"conversational"
] | null | 2024-12-06T02:36:13Z |
---
base_model: qingy2024/Fusion-14B-Instruct
language:
- zho
- eng
- fra
- spa
- por
- deu
- ita
- rus
- jpn
- kor
- vie
- tha
- ara
library_name: transformers
quantized_by: mradermacher
tags:
- mergekit
- merge
---
## About
<!-- ### quantize_version: 2 -->
<!-- ### output_tensor_quantised: 1 -->
<!-- ### convert_type: hf -->
<!-- ### vocab_type: -->
<!-- ### tags: nicoboss -->
static quants of https://huggingface.co/qingy2024/Fusion-14B-Instruct
<!-- provided-files -->
weighted/imatrix quants are available at https://huggingface.co/mradermacher/Fusion-14B-Instruct-i1-GGUF
## Usage
If you are unsure how to use GGUF files, refer to one of [TheBloke's
READMEs](https://huggingface.co/TheBloke/KafkaLM-70B-German-V0.1-GGUF) for
more details, including on how to concatenate multi-part files.
## Provided Quants
(sorted by size, not necessarily quality. IQ-quants are often preferable over similar sized non-IQ quants)
| Link | Type | Size/GB | Notes |
|:-----|:-----|--------:|:------|
| [GGUF](https://huggingface.co/mradermacher/Fusion-14B-Instruct-GGUF/resolve/main/Fusion-14B-Instruct.Q2_K.gguf) | Q2_K | 5.9 | |
| [GGUF](https://huggingface.co/mradermacher/Fusion-14B-Instruct-GGUF/resolve/main/Fusion-14B-Instruct.Q3_K_S.gguf) | Q3_K_S | 6.8 | |
| [GGUF](https://huggingface.co/mradermacher/Fusion-14B-Instruct-GGUF/resolve/main/Fusion-14B-Instruct.Q3_K_M.gguf) | Q3_K_M | 7.4 | lower quality |
| [GGUF](https://huggingface.co/mradermacher/Fusion-14B-Instruct-GGUF/resolve/main/Fusion-14B-Instruct.Q3_K_L.gguf) | Q3_K_L | 8.0 | |
| [GGUF](https://huggingface.co/mradermacher/Fusion-14B-Instruct-GGUF/resolve/main/Fusion-14B-Instruct.IQ4_XS.gguf) | IQ4_XS | 8.3 | |
| [GGUF](https://huggingface.co/mradermacher/Fusion-14B-Instruct-GGUF/resolve/main/Fusion-14B-Instruct.Q4_0_4_4.gguf) | Q4_0_4_4 | 8.6 | fast on arm, low quality |
| [GGUF](https://huggingface.co/mradermacher/Fusion-14B-Instruct-GGUF/resolve/main/Fusion-14B-Instruct.Q4_K_S.gguf) | Q4_K_S | 8.7 | fast, recommended |
| [GGUF](https://huggingface.co/mradermacher/Fusion-14B-Instruct-GGUF/resolve/main/Fusion-14B-Instruct.Q4_K_M.gguf) | Q4_K_M | 9.1 | fast, recommended |
| [GGUF](https://huggingface.co/mradermacher/Fusion-14B-Instruct-GGUF/resolve/main/Fusion-14B-Instruct.Q5_K_S.gguf) | Q5_K_S | 10.4 | |
| [GGUF](https://huggingface.co/mradermacher/Fusion-14B-Instruct-GGUF/resolve/main/Fusion-14B-Instruct.Q5_K_M.gguf) | Q5_K_M | 10.6 | |
| [GGUF](https://huggingface.co/mradermacher/Fusion-14B-Instruct-GGUF/resolve/main/Fusion-14B-Instruct.Q6_K.gguf) | Q6_K | 12.2 | very good quality |
| [GGUF](https://huggingface.co/mradermacher/Fusion-14B-Instruct-GGUF/resolve/main/Fusion-14B-Instruct.Q8_0.gguf) | Q8_0 | 15.8 | fast, best quality |
Here is a handy graph by ikawrakow comparing some lower-quality quant
types (lower is better):
And here are Artefact2's thoughts on the matter:
https://gist.github.com/Artefact2/b5f810600771265fc1e39442288e8ec9
## FAQ / Model Request
See https://huggingface.co/mradermacher/model_requests for some answers to
questions you might have and/or if you want some other model quantized.
## Thanks
I thank my company, [nethype GmbH](https://www.nethype.de/), for letting
me use its servers and providing upgrades to my workstation to enable
this work in my free time. Additional thanks to [@nicoboss](https://huggingface.co/nicoboss) for giving me access to his private supercomputer, enabling me to provide many more imatrix quants, at much higher quality, than I would otherwise be able to.
<!-- end -->
|
mradermacher/Fusion-14B-Instruct-i1-GGUF
|
mradermacher
| 2025-04-28T03:26:55Z | 25 | 1 |
transformers
|
[
"transformers",
"gguf",
"mergekit",
"merge",
"zho",
"eng",
"fra",
"spa",
"por",
"deu",
"ita",
"rus",
"jpn",
"kor",
"vie",
"tha",
"ara",
"base_model:qingy2024/Fusion-14B-Instruct",
"base_model:quantized:qingy2024/Fusion-14B-Instruct",
"endpoints_compatible",
"region:us",
"imatrix",
"conversational"
] | null | 2024-12-06T03:26:12Z |
---
base_model: qingy2024/Fusion-14B-Instruct
language:
- zho
- eng
- fra
- spa
- por
- deu
- ita
- rus
- jpn
- kor
- vie
- tha
- ara
library_name: transformers
quantized_by: mradermacher
tags:
- mergekit
- merge
---
## About
<!-- ### quantize_version: 2 -->
<!-- ### output_tensor_quantised: 1 -->
<!-- ### convert_type: hf -->
<!-- ### vocab_type: -->
<!-- ### tags: nicoboss -->
weighted/imatrix quants of https://huggingface.co/qingy2024/Fusion-14B-Instruct
<!-- provided-files -->
static quants are available at https://huggingface.co/mradermacher/Fusion-14B-Instruct-GGUF
## Usage
If you are unsure how to use GGUF files, refer to one of [TheBloke's
READMEs](https://huggingface.co/TheBloke/KafkaLM-70B-German-V0.1-GGUF) for
more details, including on how to concatenate multi-part files.
## Provided Quants
(sorted by size, not necessarily quality. IQ-quants are often preferable over similar sized non-IQ quants)
| Link | Type | Size/GB | Notes |
|:-----|:-----|--------:|:------|
| [GGUF](https://huggingface.co/mradermacher/Fusion-14B-Instruct-i1-GGUF/resolve/main/Fusion-14B-Instruct.i1-IQ1_S.gguf) | i1-IQ1_S | 3.7 | for the desperate |
| [GGUF](https://huggingface.co/mradermacher/Fusion-14B-Instruct-i1-GGUF/resolve/main/Fusion-14B-Instruct.i1-IQ1_M.gguf) | i1-IQ1_M | 4.0 | mostly desperate |
| [GGUF](https://huggingface.co/mradermacher/Fusion-14B-Instruct-i1-GGUF/resolve/main/Fusion-14B-Instruct.i1-IQ2_XXS.gguf) | i1-IQ2_XXS | 4.4 | |
| [GGUF](https://huggingface.co/mradermacher/Fusion-14B-Instruct-i1-GGUF/resolve/main/Fusion-14B-Instruct.i1-IQ2_XS.gguf) | i1-IQ2_XS | 4.8 | |
| [GGUF](https://huggingface.co/mradermacher/Fusion-14B-Instruct-i1-GGUF/resolve/main/Fusion-14B-Instruct.i1-IQ2_S.gguf) | i1-IQ2_S | 5.1 | |
| [GGUF](https://huggingface.co/mradermacher/Fusion-14B-Instruct-i1-GGUF/resolve/main/Fusion-14B-Instruct.i1-IQ2_M.gguf) | i1-IQ2_M | 5.5 | |
| [GGUF](https://huggingface.co/mradermacher/Fusion-14B-Instruct-i1-GGUF/resolve/main/Fusion-14B-Instruct.i1-Q2_K_S.gguf) | i1-Q2_K_S | 5.5 | very low quality |
| [GGUF](https://huggingface.co/mradermacher/Fusion-14B-Instruct-i1-GGUF/resolve/main/Fusion-14B-Instruct.i1-Q2_K.gguf) | i1-Q2_K | 5.9 | IQ3_XXS probably better |
| [GGUF](https://huggingface.co/mradermacher/Fusion-14B-Instruct-i1-GGUF/resolve/main/Fusion-14B-Instruct.i1-IQ3_XXS.gguf) | i1-IQ3_XXS | 6.0 | lower quality |
| [GGUF](https://huggingface.co/mradermacher/Fusion-14B-Instruct-i1-GGUF/resolve/main/Fusion-14B-Instruct.i1-IQ3_XS.gguf) | i1-IQ3_XS | 6.5 | |
| [GGUF](https://huggingface.co/mradermacher/Fusion-14B-Instruct-i1-GGUF/resolve/main/Fusion-14B-Instruct.i1-Q3_K_S.gguf) | i1-Q3_K_S | 6.8 | IQ3_XS probably better |
| [GGUF](https://huggingface.co/mradermacher/Fusion-14B-Instruct-i1-GGUF/resolve/main/Fusion-14B-Instruct.i1-IQ3_S.gguf) | i1-IQ3_S | 6.8 | beats Q3_K* |
| [GGUF](https://huggingface.co/mradermacher/Fusion-14B-Instruct-i1-GGUF/resolve/main/Fusion-14B-Instruct.i1-IQ3_M.gguf) | i1-IQ3_M | 7.0 | |
| [GGUF](https://huggingface.co/mradermacher/Fusion-14B-Instruct-i1-GGUF/resolve/main/Fusion-14B-Instruct.i1-Q3_K_M.gguf) | i1-Q3_K_M | 7.4 | IQ3_S probably better |
| [GGUF](https://huggingface.co/mradermacher/Fusion-14B-Instruct-i1-GGUF/resolve/main/Fusion-14B-Instruct.i1-Q3_K_L.gguf) | i1-Q3_K_L | 8.0 | IQ3_M probably better |
| [GGUF](https://huggingface.co/mradermacher/Fusion-14B-Instruct-i1-GGUF/resolve/main/Fusion-14B-Instruct.i1-IQ4_XS.gguf) | i1-IQ4_XS | 8.2 | |
| [GGUF](https://huggingface.co/mradermacher/Fusion-14B-Instruct-i1-GGUF/resolve/main/Fusion-14B-Instruct.i1-Q4_0_4_4.gguf) | i1-Q4_0_4_4 | 8.6 | fast on arm, low quality |
| [GGUF](https://huggingface.co/mradermacher/Fusion-14B-Instruct-i1-GGUF/resolve/main/Fusion-14B-Instruct.i1-Q4_0_4_8.gguf) | i1-Q4_0_4_8 | 8.6 | fast on arm+i8mm, low quality |
| [GGUF](https://huggingface.co/mradermacher/Fusion-14B-Instruct-i1-GGUF/resolve/main/Fusion-14B-Instruct.i1-Q4_0_8_8.gguf) | i1-Q4_0_8_8 | 8.6 | fast on arm+sve, low quality |
| [GGUF](https://huggingface.co/mradermacher/Fusion-14B-Instruct-i1-GGUF/resolve/main/Fusion-14B-Instruct.i1-Q4_0.gguf) | i1-Q4_0 | 8.6 | fast, low quality |
| [GGUF](https://huggingface.co/mradermacher/Fusion-14B-Instruct-i1-GGUF/resolve/main/Fusion-14B-Instruct.i1-Q4_K_S.gguf) | i1-Q4_K_S | 8.7 | optimal size/speed/quality |
| [GGUF](https://huggingface.co/mradermacher/Fusion-14B-Instruct-i1-GGUF/resolve/main/Fusion-14B-Instruct.i1-Q4_K_M.gguf) | i1-Q4_K_M | 9.1 | fast, recommended |
| [GGUF](https://huggingface.co/mradermacher/Fusion-14B-Instruct-i1-GGUF/resolve/main/Fusion-14B-Instruct.i1-Q5_K_S.gguf) | i1-Q5_K_S | 10.4 | |
| [GGUF](https://huggingface.co/mradermacher/Fusion-14B-Instruct-i1-GGUF/resolve/main/Fusion-14B-Instruct.i1-Q5_K_M.gguf) | i1-Q5_K_M | 10.6 | |
| [GGUF](https://huggingface.co/mradermacher/Fusion-14B-Instruct-i1-GGUF/resolve/main/Fusion-14B-Instruct.i1-Q6_K.gguf) | i1-Q6_K | 12.2 | practically like static Q6_K |
Here is a handy graph by ikawrakow comparing some lower-quality quant
types (lower is better):
And here are Artefact2's thoughts on the matter:
https://gist.github.com/Artefact2/b5f810600771265fc1e39442288e8ec9
## FAQ / Model Request
See https://huggingface.co/mradermacher/model_requests for some answers to
questions you might have and/or if you want some other model quantized.
## Thanks
I thank my company, [nethype GmbH](https://www.nethype.de/), for letting
me use its servers and providing upgrades to my workstation to enable
this work in my free time. Additional thanks to [@nicoboss](https://huggingface.co/nicoboss) for giving me access to his private supercomputer, enabling me to provide many more imatrix quants, at much higher quality, than I would otherwise be able to.
<!-- end -->
|
Jonjew/LoriSaunders
|
Jonjew
| 2025-04-28T03:16:40Z | 0 | 0 |
diffusers
|
[
"diffusers",
"text-to-image",
"lora",
"template:diffusion-lora",
"base_model:black-forest-labs/FLUX.1-dev",
"base_model:adapter:black-forest-labs/FLUX.1-dev",
"license:unknown",
"region:us"
] |
text-to-image
| 2025-04-28T03:16:34Z |
---
tags:
- text-to-image
- lora
- diffusers
- template:diffusion-lora
widget:
- text: Lori
output:
url: images/bobbyjo.png
base_model: black-forest-labs/FLUX.1-dev
instance_prompt: Lori
license: unknown
---
# Lori Saunders (2nd Bobbie Jo Bradley from Petticoat Junction) by nobodyishome
<Gallery />
## Model description
FROM https://civitai.com/models/1510312/lori-saunders-2nd-bobbie-jo-bradley
Please support the creator by Liking at the page above
Trigger Lori
## Trigger words
You should use `Lori` to trigger the image generation.
## Download model
Weights for this model are available in Safetensors format.
[Download](/Jonjew/LoriSaunders/tree/main) them in the Files & versions tab.
|
Subsets and Splits
Filtered Qwen2.5 Distill Models
Identifies specific configurations of models by filtering cards that contain 'distill', 'qwen2.5', '7b' while excluding certain base models and incorrect model ID patterns, uncovering unique model variants.
Filtered Model Cards Count
Finds the count of entries with specific card details that include 'distill', 'qwen2.5', '7b' but exclude certain base models, revealing valuable insights about the dataset's content distribution.
Filtered Distill Qwen 7B Models
Filters for specific card entries containing 'distill', 'qwen', and '7b', excluding certain strings and patterns, to identify relevant model configurations.
Filtered Qwen-7b Model Cards
The query performs a detailed filtering based on specific keywords and excludes certain entries, which could be useful for identifying a specific subset of cards but does not provide deeper insights or trends.
Filtered Qwen 7B Model Cards
The query filters for specific terms related to "distilled" or "distill", "qwen", and "7b" in the 'card' column but excludes certain base models, providing a limited set of entries for further inspection.
Qwen 7B Distilled Models
The query provides a basic filtering of records to find specific card names that include keywords related to distilled Qwen 7b models, excluding a particular base model, which gives limited insight but helps in focusing on relevant entries.
Qwen 7B Distilled Model Cards
The query filters data based on specific keywords in the modelId and card fields, providing limited insight primarily useful for locating specific entries rather than revealing broad patterns or trends.
Qwen 7B Distilled Models
Finds all entries containing the terms 'distilled', 'qwen', and '7b' in a case-insensitive manner, providing a filtered set of records but without deeper analysis.
Distilled Qwen 7B Models
The query filters for specific model IDs containing 'distilled', 'qwen', and '7b', providing a basic retrieval of relevant entries but without deeper analysis or insight.
Filtered Model Cards with Distill Qwen2.
Filters and retrieves records containing specific keywords in the card description while excluding certain phrases, providing a basic count of relevant entries.
Filtered Model Cards with Distill Qwen 7
The query filters specific variations of card descriptions containing 'distill', 'qwen', and '7b' while excluding a particular base model, providing limited but specific data retrieval.
Distill Qwen 7B Model Cards
The query filters and retrieves rows where the 'card' column contains specific keywords ('distill', 'qwen', and '7b'), providing a basic filter result that can help in identifying specific entries.