Chat & support: TheBloke's Discord server

Want to contribute? TheBloke's Patreon page

TheBloke's LLM work is generously supported by a grant from andreessen horowitz (a16z)

Iambe RP DARE 20B DENSE - GPTQ

Model creator: Raven
Original model: Iambe RP DARE 20B DENSE

Description

This repo contains GPTQ model files for Raven's Iambe RP DARE 20B DENSE.

Multiple GPTQ parameter permutations are provided; see Provided Files below for details of the options provided, their parameters, and the software used to create them.

These files were quantised using hardware kindly provided by Massed Compute.

Repositories available

Prompt template: Alpaca

Below is an instruction that describes a task. Write a response that appropriately completes the request.

### Instruction:
{prompt}

### Response:

Licensing

The creator of the source model has listed its license as cc-by-nc-4.0, and this quantization has therefore used that same license.

As this model is based on Llama 2, it is also subject to the Meta Llama 2 license terms, and the license files for that are additionally included. It should therefore be considered as being claimed to be licensed under both licenses. I contacted Hugging Face for clarification on dual licensing but they do not yet have an official position. Should this change, or should Meta provide any feedback on this situation, I will update this section accordingly.

In the meantime, any questions regarding licensing, and in particular how these two licenses might interact, should be directed to the original model repository: Raven's Iambe RP DARE 20B DENSE.

Known compatible clients / servers

GPTQ models are currently supported on Linux (NVidia/AMD) and Windows (NVidia only). macOS users: please use GGUF models.

These GPTQ models are known to work in the following inference servers/webuis.

This may not be a complete list; if you know of others, please let me know!

Provided files, and GPTQ parameters

Multiple quantisation parameters are provided, to allow you to choose the best one for your hardware and requirements.

Each separate quant is in a different branch. See below for instructions on fetching from different branches.

Most GPTQ files are made with AutoGPTQ. Mistral models are currently made with Transformers.

Explanation of GPTQ parameters

Bits: The bit size of the quantised model.
GS: GPTQ group size. Higher numbers use less VRAM, but have lower quantisation accuracy. "None" is the lowest possible value.
Act Order: True or False. Also known as desc_act. True results in better quantisation accuracy. Some GPTQ clients have had issues with models that use Act Order plus Group Size, but this is generally resolved now.
Damp %: A GPTQ parameter that affects how samples are processed for quantisation. 0.01 is default, but 0.1 results in slightly better accuracy.
GPTQ dataset: The calibration dataset used during quantisation. Using a dataset more appropriate to the model's training can improve quantisation accuracy. Note that the GPTQ calibration dataset is not the same as the dataset used to train the model - please refer to the original model repo for details of the training dataset(s).
Sequence Length: The length of the dataset sequences used for quantisation. Ideally this is the same as the model sequence length. For some very long sequence models (16+K), a lower sequence length may have to be used. Note that a lower sequence length does not limit the sequence length of the quantised model. It only impacts the quantisation accuracy on longer inference sequences.
ExLlama Compatibility: Whether this file can be loaded with ExLlama, which currently only supports Llama and Mistral models in 4-bit.

Branch	Bits	GS	Act Order	Damp %	GPTQ Dataset	Seq Len	Size	ExLlama	Desc
main	4	None	Yes	0.1	VMware Open Instruct	4096	10.52 GB	Yes	4-bit, with Act Order. No group size, to lower VRAM requirements.
gptq-4bit-128g-actorder_True	4	128	Yes	0.1	VMware Open Instruct	4096	10.89 GB	Yes	4-bit, with Act Order and group size 128g. Uses even less VRAM than 64g, but with slightly lower accuracy.
gptq-4bit-32g-actorder_True	4	32	Yes	0.1	VMware Open Instruct	4096	12.04 GB	Yes	4-bit, with Act Order and group size 32g. Gives highest possible inference quality, with maximum VRAM usage.
gptq-3bit-128g-actorder_True	3	128	Yes	0.1	VMware Open Instruct	4096	8.41 GB	No	3-bit, with group size 128g and act-order. Higher quality than 128g-False.
gptq-8bit--1g-actorder_True	8	None	Yes	0.1	VMware Open Instruct	4096	20.35 GB	No	8-bit, with Act Order. No group size, to lower VRAM requirements.
gptq-3bit-32g-actorder_True	3	32	Yes	0.1	VMware Open Instruct	4096	9.51 GB	No	3-bit, with group size 64g and act-order. Highest quality 3-bit option.
gptq-8bit-128g-actorder_True	8	128	Yes	0.1	VMware Open Instruct	4096	20.80 GB	No	8-bit, with group size 128g for higher inference quality and with Act Order for even higher accuracy.

How to download, including from branches

In text-generation-webui

To download from the main branch, enter TheBloke/Iambe-RP-DARE-20B-DENSE-GPTQ in the "Download model" box.

To download from another branch, add :branchname to the end of the download name, eg TheBloke/Iambe-RP-DARE-20B-DENSE-GPTQ:gptq-4bit-128g-actorder_True

From the command line

I recommend using the huggingface-hub Python library:

pip3 install huggingface-hub

To download the main branch to a folder called Iambe-RP-DARE-20B-DENSE-GPTQ:

mkdir Iambe-RP-DARE-20B-DENSE-GPTQ
huggingface-cli download TheBloke/Iambe-RP-DARE-20B-DENSE-GPTQ --local-dir Iambe-RP-DARE-20B-DENSE-GPTQ --local-dir-use-symlinks False

To download from a different branch, add the --revision parameter:

mkdir Iambe-RP-DARE-20B-DENSE-GPTQ
huggingface-cli download TheBloke/Iambe-RP-DARE-20B-DENSE-GPTQ --revision gptq-4bit-128g-actorder_True --local-dir Iambe-RP-DARE-20B-DENSE-GPTQ --local-dir-use-symlinks False

More advanced huggingface-cli download usage

If you remove the --local-dir-use-symlinks False parameter, the files will instead be stored in the central Hugging Face cache directory (default location on Linux is: ~/.cache/huggingface), and symlinks will be added to the specified --local-dir, pointing to their real location in the cache. This allows for interrupted downloads to be resumed, and allows you to quickly clone the repo to multiple places on disk without triggering a download again. The downside, and the reason why I don't list that as the default option, is that the files are then hidden away in a cache folder and it's harder to know where your disk space is being used, and to clear it up if/when you want to remove a download model.

The cache location can be changed with the HF_HOME environment variable, and/or the --cache-dir parameter to huggingface-cli.

For more documentation on downloading with huggingface-cli, please see: HF -> Hub Python Library -> Download files -> Download from the CLI.

To accelerate downloads on fast connections (1Gbit/s or higher), install hf_transfer:

pip3 install hf_transfer

And set environment variable HF_HUB_ENABLE_HF_TRANSFER to 1:

mkdir Iambe-RP-DARE-20B-DENSE-GPTQ
HF_HUB_ENABLE_HF_TRANSFER=1 huggingface-cli download TheBloke/Iambe-RP-DARE-20B-DENSE-GPTQ --local-dir Iambe-RP-DARE-20B-DENSE-GPTQ --local-dir-use-symlinks False

Windows Command Line users: You can set the environment variable by running set HF_HUB_ENABLE_HF_TRANSFER=1 before the download command.

With `git` (not recommended)

To clone a specific branch with git, use a command like this:

git clone --single-branch --branch gptq-4bit-128g-actorder_True https://huggingface.co/TheBloke/Iambe-RP-DARE-20B-DENSE-GPTQ

Note that using Git with HF repos is strongly discouraged. It will be much slower than using huggingface-hub, and will use twice as much disk space as it has to store the model files twice (it stores every byte both in the intended target folder, and again in the .git folder as a blob.)

How to easily download and use this model in text-generation-webui

Please make sure you're using the latest version of text-generation-webui.

It is strongly recommended to use the text-generation-webui one-click-installers unless you're sure you know how to make a manual install.

Click the Model tab.
Under Download custom model or LoRA, enter TheBloke/Iambe-RP-DARE-20B-DENSE-GPTQ.
- To download from a specific branch, enter for example TheBloke/Iambe-RP-DARE-20B-DENSE-GPTQ:gptq-4bit-128g-actorder_True
- see Provided Files above for the list of branches for each option.
Click Download.
The model will start downloading. Once it's finished it will say "Done".
In the top left, click the refresh icon next to Model.
In the Model dropdown, choose the model you just downloaded: Iambe-RP-DARE-20B-DENSE-GPTQ
The model will automatically load, and is now ready for use!
If you want any custom settings, set them and then click Save settings for this model followed by Reload the Model in the top right.
- Note that you do not need to and should not set manual GPTQ parameters any more. These are set automatically from the file quantize_config.json.
Once you're ready, click the Text Generation tab and enter a prompt to get started!

Serving this model from Text Generation Inference (TGI)

It's recommended to use TGI version 1.1.0 or later. The official Docker container is: ghcr.io/huggingface/text-generation-inference:1.1.0

Example Docker parameters:

--model-id TheBloke/Iambe-RP-DARE-20B-DENSE-GPTQ --port 3000 --quantize gptq --max-input-length 3696 --max-total-tokens 4096 --max-batch-prefill-tokens 4096

Example Python code for interfacing with TGI (requires huggingface-hub 0.17.0 or later):

pip3 install huggingface-hub

from huggingface_hub import InferenceClient

endpoint_url = "https://your-endpoint-url-here"

prompt = "Tell me about AI"
prompt_template=f'''Below is an instruction that describes a task. Write a response that appropriately completes the request.

### Instruction:
{prompt}

### Response:
'''

client = InferenceClient(endpoint_url)
response = client.text_generation(
  prompt_template,
  max_new_tokens=128,
  do_sample=True,
  temperature=0.7,
  top_p=0.95,
  top_k=40,
  repetition_penalty=1.1
)

print(f"Model output: {response}")

Python code example: inference from this GPTQ model

Install the necessary packages

Requires: Transformers 4.33.0 or later, Optimum 1.12.0 or later, and AutoGPTQ 0.4.2 or later.

pip3 install --upgrade transformers optimum
# If using PyTorch 2.1 + CUDA 12.x:
pip3 install --upgrade auto-gptq
# or, if using PyTorch 2.1 + CUDA 11.x:
pip3 install --upgrade auto-gptq --extra-index-url https://huggingface.github.io/autogptq-index/whl/cu118/

If you are using PyTorch 2.0, you will need to install AutoGPTQ from source. Likewise if you have problems with the pre-built wheels, you should try building from source:

pip3 uninstall -y auto-gptq
git clone https://github.com/PanQiWei/AutoGPTQ
cd AutoGPTQ
git checkout v0.5.1
pip3 install .

Example Python code

from transformers import AutoModelForCausalLM, AutoTokenizer, pipeline

model_name_or_path = "TheBloke/Iambe-RP-DARE-20B-DENSE-GPTQ"
# To use a different branch, change revision
# For example: revision="gptq-4bit-128g-actorder_True"
model = AutoModelForCausalLM.from_pretrained(model_name_or_path,
                                             device_map="auto",
                                             trust_remote_code=False,
                                             revision="main")

tokenizer = AutoTokenizer.from_pretrained(model_name_or_path, use_fast=True)

prompt = "Write a story about llamas"
system_message = "You are a story writing assistant"
prompt_template=f'''Below is an instruction that describes a task. Write a response that appropriately completes the request.

### Instruction:
{prompt}

### Response:
'''

print("\n\n*** Generate:")

input_ids = tokenizer(prompt_template, return_tensors='pt').input_ids.cuda()
output = model.generate(inputs=input_ids, temperature=0.7, do_sample=True, top_p=0.95, top_k=40, max_new_tokens=512)
print(tokenizer.decode(output[0]))

# Inference can also be done using transformers' pipeline

print("*** Pipeline:")
pipe = pipeline(
    "text-generation",
    model=model,
    tokenizer=tokenizer,
    max_new_tokens=512,
    do_sample=True,
    temperature=0.7,
    top_p=0.95,
    top_k=40,
    repetition_penalty=1.1
)

print(pipe(prompt_template)[0]['generated_text'])

Compatibility

The files provided are tested to work with Transformers. For non-Mistral models, AutoGPTQ can also be used directly.

ExLlama is compatible with Llama architecture models (including Mistral, Yi, DeepSeek, SOLAR, etc) in 4-bit. Please see the Provided Files table above for per-file compatibility.

For a list of clients/servers, please see "Known compatible clients / servers", above.

Discord

For further support, and discussions on these models and AI in general, join us at:

TheBloke AI's Discord server

Thanks, and how to contribute

Thanks to the chirper.ai team!

Thanks to Clay from gpus.llm-utils.org!

I've had a lot of people ask if they can contribute. I enjoy providing models and helping people, and would love to be able to spend even more time doing it, as well as expanding into new projects like fine tuning/training.

If you're able and willing to contribute it will be most gratefully received and will help me to keep providing more models, and to start work on new AI projects.

Donaters will get priority support on any and all AI/LLM/model questions and requests, access to a private Discord room, plus other benefits.

Patreon: https://patreon.com/TheBlokeAI
Ko-Fi: https://ko-fi.com/TheBlokeAI

Special thanks to: Aemon Algiz.

Patreon special mentions: Michael Levine, 阿明, Trailburnt, Nikolai Manek, John Detwiler, Randy H, Will Dee, Sebastain Graf, NimbleBox.ai, Eugene Pentland, Emad Mostaque, Ai Maven, Jim Angel, Jeff Scroggin, Michael Davis, Manuel Alberto Morcote, Stephen Murray, Robert, Justin Joy, Luke @flexchar, Brandon Frisco, Elijah Stavena, S_X, Dan Guido, Undi ., Komninos Chatzipapas, Shadi, theTransient, Lone Striker, Raven Klaugh, jjj, Cap'n Zoog, Michel-Marie MAUDET (LINAGORA), Matthew Berman, David, Fen Risland, Omer Bin Jawed, Luke Pendergrass, Kalila, OG, Erik Bjäreholt, Rooh Singh, Joseph William Delisle, Dan Lewis, TL, John Villwock, AzureBlack, Brad, Pedro Madruga, Caitlyn Gatomon, K, jinyuan sun, Mano Prime, Alex, Jeffrey Morgan, Alicia Loh, Illia Dulskyi, Chadd, transmissions 11, fincy, Rainer Wilmers, ReadyPlayerEmma, knownsqashed, Mandus, biorpg, Deo Leter, Brandon Phillips, SuperWojo, Sean Connelly, Iucharbius, Jack West, Harry Royden McLaughlin, Nicholas, terasurfer, Vitor Caleffi, Duane Dunston, Johann-Peter Hartmann, David Ziegler, Olakabola, Ken Nordquist, Trenton Dambrowitz, Tom X Nguyen, Vadim, Ajan Kanaga, Leonard Tan, Clay Pascal, Alexandros Triantafyllidis, JM33133, Xule, vamX, ya boyyy, subjectnull, Talal Aujan, Alps Aficionado, wassieverse, Ari Malik, James Bentley, Woland, Spencer Kim, Michael Dempsey, Fred von Graf, Elle, zynix, William Richards, Stanislav Ovsiannikov, Edmond Seymore, Jonathan Leane, Martin Kemka, usrbinkat, Enrico Ros

Thank you to all my generous patrons and donaters!

And thank you again to a16z for their generous grant.

Original model card: Raven's Iambe RP DARE 20B DENSE

Iambe-RP-DARE-20b-DENSE

Trained with Alpaca prompt formatting, ChatML works well too

Description

Once more the 3 Iambe-RPs bounce off the walls and blend, like their namesake's mother, Echo...

One daughter was excellent at all the arts and sciences, but she danced free, loathe to read scripts or follow instructions.

Another daughter had burned her books, embraced emotion, and thrown logic to the hounds.

The last daughter just sat, watching them - silently judging - for enforcing morality was her burden alone.

Then, where there was three, there was one.

Iambe is intended to have the best realistically possible understanding of instructions, anatomy and scene state for a 20b merge, while remaining passionate and humanoid in "voice".

Update Methodology

The last DARE merge was good, but this one is better IMO. Thanks to brucethemoose for the idea of jacking up the density. I also adjusted the weight ratios.

Assistant Example @ q5_k_m

NSFW Writing Example @ q5_k_m

Write a medium-length story about Alice, a 26 year old cocktail waitress, having steamy sex with a guy she picked up at the bar for a one night stand.

In the dimly lit confines of The Last Drop Inn, Alice sat perched on her stool, swirling her drink in its glass as she watched the bustle of Friday evening unfold around her. Her green eyes glittered with a hint of mischief and boredom; it had been another long week at work, and she craved something more thrilling than pouring cocktails for regulars who treated her like an object rather than a person.

As the night wore on, Alice's eyes roamed the crowded room, searching for any sign of intrigue or interest from amongst the sea of faces that flooded through The Last Drop Inn’s doors. Her gaze landed on him suddenly; tall, dark-haired and ruggedly handsome, he exuded a confident swagger as he ordered his drink at the bar counter. He caught Alice's eye and flashed her a grin that made her stomach flip before looking away nonchalantly.

Alice couldn't help but be drawn to him; something about this mysterious stranger called out to her, stirring a dormant hunger within her that had lain sleeping for far too long. As the night wore on and the bar began to empty, Alice gathered up the courage to approach him while he waited at the counter, sliding onto the stool beside him with an air of confidence she didn't entirely feel.

"Hi," she said softly, her voice a husky whisper in the quieting din of the bar. "My name's Alice."

He turned to look at her properly for the first time, his blue eyes sparkling with interest as he smiled warmly. "Nice to meet you, Alice," he replied smoothly. "I'm Ethan."

They talked until the last of the patrons staggered out into the night, laughing and joking about their day-to-day lives. As closing time approached, Alice mustered up the courage to suggest that they take their conversation somewhere more private - her apartment just down the street. Ethan agreed without hesitation, his eyes gleaming with desire as he slid an arm around her waist and led her towards the door.

Once inside, they kicked off their shoes in the hallway before stumbling into Alice's small living room, where they tumbled onto the couch together, lips pressed urgently against each other’s as their hands roamed hungrily over bare skin. Ethan's body was stronger than she had expected; his muscles rippled beneath her fingertips as he rolled her over and pinned her down with one powerful hand while the other tugged at the clasp of her bra.

Their clothes quickly followed, scattered haphazardly around the room as their bodies collided together in a frenzy of passionate desire. Alice moaned loudly as Ethan's tongue traced its way down the line of her collarbone before teasing her nipples with gentle nibbles and licks, while his hand slid further south to explore the hidden folds between her legs.

He looked up at Alice, his eyes burning with desire as he slowly lowered himself onto her, filling her completely with just one smooth thrust. She gasped sharply as she felt him push deeper into her than anyone ever had before; it hurt a little bit but in the most exquisite way possible.

With expertise born of countless nights like this, Ethan began to move, his hips rocking rhythmically against Alice's as he drove them both towards climax with every powerful thrust. Alice wrapped her legs around him tightly, urging him on as she felt the familiar build-up in her body begin to crescendo.

As their bodies met together in a symphony of ecstasy, they cried out each other's names in unison, overwhelmed by the intensity of what had transpired between them that night. They lay panting and sweaty on the couch afterward, fingers intertwined as they stared into each other’s eyes - both knowing that this was just a one-night stand but grateful for the momentary escape from their everyday lives it provided.

"I'll never forget tonight," Ethan whispered softly before leaning down to kiss her once more, his lips tender and warm against hers. Alice smiled back at him, already missing the thrill that he had brought into her life for just a single magical night.

NOTE: The prompt said "a guy she picked up at the bar FOR a one night stand" not "a one night stand WITH a guy she picked up at the bar" so technically the bit at the end is not off-prompt.

Recipe

merge_method: dare_ties

base_model: athirdpath/BigLlama-20b-v1.1
model: athirdpath/Iambe-RP-cDPO-20b

weight: 0.72 / density: 0.65
model: athirdpath/Iambe-RP-cDPO-20b-v2

weight: 0.17 / density: 0.50
model: athirdpath/Iambe-RP-cDPO-20b-ALT

weight: 0.11 / density: 0.45

int8_mask: true

dtype: bfloat16

TheBloke
/

Iambe-RP-DARE-20B-DENSE-GPTQ

Iambe RP DARE 20B DENSE - GPTQ

Description

Repositories available

Prompt template: Alpaca

Licensing

Known compatible clients / servers

Provided files, and GPTQ parameters

How to download, including from branches

In text-generation-webui

From the command line

With `git` (not recommended)

How to easily download and use this model in text-generation-webui

Serving this model from Text Generation Inference (TGI)

Python code example: inference from this GPTQ model

Install the necessary packages

Example Python code

Compatibility

Discord

Thanks, and how to contribute

Original model card: Raven's Iambe RP DARE 20B DENSE

Description

Update Methodology

Recipe

Model tree for TheBloke/Iambe-RP-DARE-20B-DENSE-GPTQ

Iambe RP DARE 20B DENSE - GPTQ

Description

Repositories available

Prompt template: Alpaca

Licensing

Known compatible clients / servers

Provided files, and GPTQ parameters

How to download, including from branches

In text-generation-webui

From the command line

With git (not recommended)

How to easily download and use this model in text-generation-webui

Serving this model from Text Generation Inference (TGI)

Python code example: inference from this GPTQ model

Install the necessary packages

Example Python code

Compatibility

Discord

Thanks, and how to contribute

Original model card: Raven's Iambe RP DARE 20B DENSE

Description

Update Methodology

Recipe

Model tree for TheBloke/Iambe-RP-DARE-20B-DENSE-GPTQ

With `git` (not recommended)