|
|
--- |
|
|
base_model: mistralai/Mistral-7B-Instruct-v0.3 |
|
|
datasets: |
|
|
- generator |
|
|
library_name: peft |
|
|
license: apache-2.0 |
|
|
tags: |
|
|
- trl |
|
|
- sft |
|
|
- generated_from_trainer |
|
|
model-index: |
|
|
- name: mistral_7b_cosine_lr |
|
|
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. --> |
|
|
|
|
|
# mistral_7b_cosine_lr |
|
|
|
|
|
This model is a fine-tuned version of [mistralai/Mistral-7B-Instruct-v0.3](https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.3) on the generator dataset. |
|
|
It achieves the following results on the evaluation set: |
|
|
- Loss: 0.3810 |
|
|
|
|
|
## 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: 0.0002 |
|
|
- train_batch_size: 2 |
|
|
- eval_batch_size: 8 |
|
|
- seed: 42 |
|
|
- gradient_accumulation_steps: 8 |
|
|
- total_train_batch_size: 16 |
|
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
|
- lr_scheduler_type: cosine |
|
|
- lr_scheduler_warmup_ratio: 0.03 |
|
|
- lr_scheduler_warmup_steps: 15 |
|
|
- num_epochs: 3 |
|
|
|
|
|
### Training results |
|
|
|
|
|
| Training Loss | Epoch | Step | Validation Loss | |
|
|
|:-------------:|:------:|:----:|:---------------:| |
|
|
| 0.8199 | 0.0366 | 10 | 0.6013 | |
|
|
| 0.5795 | 0.0732 | 20 | 0.5157 | |
|
|
| 0.5089 | 0.1098 | 30 | 0.4825 | |
|
|
| 0.4742 | 0.1465 | 40 | 0.4628 | |
|
|
| 0.468 | 0.1831 | 50 | 0.4502 | |
|
|
| 0.4489 | 0.2197 | 60 | 0.4424 | |
|
|
| 0.4471 | 0.2563 | 70 | 0.4378 | |
|
|
| 0.4529 | 0.2929 | 80 | 0.4365 | |
|
|
| 0.4382 | 0.3295 | 90 | 0.4289 | |
|
|
| 0.4352 | 0.3661 | 100 | 0.4267 | |
|
|
| 0.4331 | 0.4027 | 110 | 0.4214 | |
|
|
| 0.4368 | 0.4394 | 120 | 0.4326 | |
|
|
| 0.4244 | 0.4760 | 130 | 0.4188 | |
|
|
| 0.4231 | 0.5126 | 140 | 0.4143 | |
|
|
| 0.418 | 0.5492 | 150 | 0.4103 | |
|
|
| 0.4118 | 0.5858 | 160 | 0.4066 | |
|
|
| 0.4151 | 0.6224 | 170 | 0.4047 | |
|
|
| 0.4093 | 0.6590 | 180 | 0.4020 | |
|
|
| 0.4066 | 0.6957 | 190 | 0.4004 | |
|
|
| 0.4072 | 0.7323 | 200 | 0.3970 | |
|
|
| 0.4002 | 0.7689 | 210 | 0.3933 | |
|
|
| 0.3968 | 0.8055 | 220 | 0.3919 | |
|
|
| 0.3967 | 0.8421 | 230 | 0.3893 | |
|
|
| 0.3901 | 0.8787 | 240 | 0.3870 | |
|
|
| 0.3895 | 0.9153 | 250 | 0.3852 | |
|
|
| 0.3929 | 0.9519 | 260 | 0.3823 | |
|
|
| 0.3854 | 0.9886 | 270 | 0.3808 | |
|
|
| 0.3524 | 1.0252 | 280 | 0.3833 | |
|
|
| 0.332 | 1.0618 | 290 | 0.3821 | |
|
|
| 0.3283 | 1.0984 | 300 | 0.3810 | |
|
|
|
|
|
|
|
|
### Framework versions |
|
|
|
|
|
- PEFT 0.13.2 |
|
|
- Transformers 4.45.2 |
|
|
- Pytorch 2.4.1+cu121 |
|
|
- Datasets 3.0.1 |
|
|
- Tokenizers 0.20.0 |
