13B-Ouroboros-AWQ

<!-- header start --> <!-- 200823 --> <div style="width: auto; margin-left: auto; margin-right: auto"> <img src="https://i.imgur.com/EBdldam.jpg" alt="TheBlokeAI" style="width: 100%; min-width: 400px; display: block; margin: auto;"> </div> <div style="display: flex; justify-content: space-between; width: 100%;"> <div style="display: flex; flex-direction: column; align-items: flex-start;"> <p style="margin-top: 0.5em; margin-bottom: 0em;"><a href="https://discord.gg/theblokeai">Chat & support: TheBloke's Discord server</a></p> </div> <div style="display: flex; flex-direction: column; align-items: flex-end;"> <p style="margin-top: 0.5em; margin-bottom: 0em;"><a href="https://www.patreon.com/TheBlokeAI">Want to contribute? TheBloke's Patreon page</a></p> </div> </div> <div style="text-align:center; margin-top: 0em; margin-bottom: 0em"><p style="margin-top: 0.25em; margin-bottom: 0em;">TheBloke's LLM work is generously supported by a grant from <a href="https://a16z.com">andreessen horowitz (a16z)</a></p></div> <hr style="margin-top: 1.0em; margin-bottom: 1.0em;"> <!-- header end -->

13B Ouroboros - AWQ

• Model creator: Caldera AI
• Original model: 13B Ouroboros

<!-- description start -->

Description

This repo contains AWQ model files for CalderaAI's 13B Ouroboros.

About AWQ

AWQ is an efficient, accurate and blazing-fast low-bit weight quantization method, currently supporting 4-bit quantization. Compared to GPTQ, it offers faster Transformers-based inference.

It is also now supported by continuous batching server vLLM, allowing use of AWQ models for high-throughput concurrent inference in multi-user server scenarios. Note that, at the time of writing, overall throughput is still lower than running vLLM with unquantised models, however using AWQ enables using much smaller GPUs which can lead to easier deployment and overall cost savings. For example, a 70B model can be run on 1 x 48GB GPU instead of 2 x 80GB.

<!-- description end --> <!-- repositories-available start -->

Repositories available

• AWQ model(s) for GPU inference.
• GPTQ models for GPU inference, with multiple quantisation parameter options.
• 2, 3, 4, 5, 6 and 8-bit GGUF models for CPU+GPU inference
• Caldera AI's original unquantised fp16 model in pytorch format, for GPU inference and for further conversions

<!-- repositories-available end --> <!-- prompt-template start -->

Prompt template: Alpaca

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

### Instruction:
{prompt}

### Response:

<!-- prompt-template end --> <!-- README_AWQ.md-provided-files start -->

Provided files and AWQ parameters

For my first release of AWQ models, I am releasing 128g models only. I will consider adding 32g as well if there is interest, and once I have done perplexity and evaluation comparisons, but at this time 32g models are still not fully tested with AutoAWQ and vLLM.

Models are released as sharded safetensors files.

<!-- README_AWQ.md-provided-files end --> <!-- README_AWQ.md-use-from-vllm start -->

Serving this model from vLLM

Documentation on installing and using vLLM can be found here.

• When using vLLM as a server, pass the --quantization awq parameter, for example:

python3 python -m vllm.entrypoints.api_server --model TheBloke/13B-Ouroboros-AWQ --quantization awq

When using vLLM from Python code, pass the quantization=awq parameter, for example:

from vllm import LLM, SamplingParams

prompts = [
    "Hello, my name is",
    "The president of the United States is",
    "The capital of France is",
    "The future of AI is",
]
sampling_params = SamplingParams(temperature=0.8, top_p=0.95)

llm = LLM(model="TheBloke/13B-Ouroboros-AWQ", quantization="awq")

outputs = llm.generate(prompts, sampling_params)

# Print the outputs.
for output in outputs:
    prompt = output.prompt
    generated_text = output.outputs[0].text
    print(f"Prompt: {prompt!r}, Generated text: {generated_text!r}")

<!-- README_AWQ.md-use-from-vllm start --> <!-- README_AWQ.md-use-from-python start -->

How to use this AWQ model from Python code

Install the necessary packages

Requires: AutoAWQ 0.0.2 or later

pip3 install autoawq

If you have problems installing AutoAWQ using the pre-built wheels, install it from source instead:

pip3 uninstall -y autoawq
git clone https://github.com/casper-hansen/AutoAWQ
cd AutoAWQ
pip3 install .

You can then try the following example code

from awq import AutoAWQForCausalLM
from transformers import AutoTokenizer

model_name_or_path = "TheBloke/13B-Ouroboros-AWQ"

# Load model
model = AutoAWQForCausalLM.from_quantized(model_name_or_path, fuse_layers=True,
                                          trust_remote_code=True, safetensors=True)
tokenizer = AutoTokenizer.from_pretrained(model_name_or_path, trust_remote_code=True)

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:

'''

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

tokens = tokenizer(
    prompt_template,
    return_tensors='pt'
).input_ids.cuda()

# Generate output
generation_output = model.generate(
    tokens,
    do_sample=True,
    temperature=0.7,
    top_p=0.95,
    top_k=40,
    max_new_tokens=512
)

print("Output: ", tokenizer.decode(generation_output[0]))

# Inference can also be done using transformers' pipeline
from transformers import 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'])

<!-- README_AWQ.md-use-from-python end --> <!-- README_AWQ.md-compatibility start -->

Compatibility

The files provided are tested to work with AutoAWQ, and vLLM.

Huggingface Text Generation Inference (TGI) is not yet compatible with AWQ, but a PR is open which should bring support soon: TGI PR #781.

<!-- README_AWQ.md-compatibility end --> <!-- footer start --> <!-- 200823 -->

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: Alicia Loh, Stephen Murray, K, Ajan Kanaga, RoA, Magnesian, Deo Leter, Olakabola, Eugene Pentland, zynix, Deep Realms, Raymond Fosdick, Elijah Stavena, Iucharbius, Erik Bjäreholt, Luis Javier Navarrete Lozano, Nicholas, theTransient, John Detwiler, alfie_i, knownsqashed, Mano Prime, Willem Michiel, Enrico Ros, LangChain4j, OG, Michael Dempsey, Pierre Kircher, Pedro Madruga, James Bentley, Thomas Belote, Luke @flexchar, Leonard Tan, Johann-Peter Hartmann, Illia Dulskyi, Fen Risland, Chadd, S_X, Jeff Scroggin, Ken Nordquist, Sean Connelly, Artur Olbinski, Swaroop Kallakuri, Jack West, Ai Maven, David Ziegler, Russ Johnson, transmissions 11, John Villwock, Alps Aficionado, Clay Pascal, Viktor Bowallius, Subspace Studios, Rainer Wilmers, Trenton Dambrowitz, vamX, Michael Levine, 준교 김, Brandon Frisco, Kalila, Trailburnt, Randy H, Talal Aujan, Nathan Dryer, Vadim, 阿明, ReadyPlayerEmma, Tiffany J. Kim, George Stoitzev, Spencer Kim, Jerry Meng, Gabriel Tamborski, Cory Kujawski, Jeffrey Morgan, Spiking Neurons AB, Edmond Seymore, Alexandros Triantafyllidis, Lone Striker, Cap'n Zoog, Nikolai Manek, danny, ya boyyy, Derek Yates, usrbinkat, Mandus, TL, Nathan LeClaire, subjectnull, Imad Khwaja, webtim, Raven Klaugh, Asp the Wyvern, Gabriel Puliatti, Caitlyn Gatomon, Joseph William Delisle, Jonathan Leane, Luke Pendergrass, SuperWojo, Sebastain Graf, Will Dee, Fred von Graf, Andrey, Dan Guido, Daniel P. Andersen, Nitin Borwankar, Elle, Vitor Caleffi, biorpg, jjj, NimbleBox.ai, Pieter, Matthew Berman, terasurfer, Michael Davis, Alex, Stanislav Ovsiannikov

Thank you to all my generous patrons and donaters!

And thank you again to a16z for their generous grant.

<!-- footer end -->

Original model card: CalderaAI's 13B Ouroboros

13B-Ouroboros

Ouroboros is an experimental model based on Meta's LLaMA [v1] 13B base model using a custom merging technique, tweaking each layer's merge % based on internal tests against the PTB dataset, scoring ~26.31 according to internal evaluation (6 samples, sequence length 1024; this testing is not empirical, it's a quick way to find near-optimum values). Testing, evaluating, and remixing this model is absolutely permissible and even encouraged (within the bounds of Meta's LLaMAv1 license agreement); the more feedback the better we can tune our process! 😊

Composition:

Ouroboros is comprised of 40 layers [LLaMAv1 13B standard] mixed at optimized ratios VS the PTB dataset for lowest perplexity score. Listed below are the paired models and ratios merged per layer.

Tier One Merge:

13B-airoboros-gpt4-1.4 > 13B-orca_mini_v2

[0.22, 0.85, 0.89, 0.98, 0.3, 0.41, 0.71, 0.83, 0.32, 0.1, 0.44, 0.6, 0.53, 0.15, 0.86, 0.79, 0.93, 0.02, 0.19, 0.82, 0.01, 0.52, 0.07, 0.27, 0.73, 0.86, 0.08, 0.67, 0.42, 0.28, 0.37, 0.08, 0.95, 0.68, 0.45, 0.08, 0.7, 0.93, 0.96, 0.43]

13B-gpt4-x-alpaca > 13B-Vicuna-cocktail

[0.65, 0.94, 0.98, 0.87, 0.28, 0.64, 0.73, 0.7, 0.95, 0.89, 0.84, 0.9, 0.59, 0.92, 0.28, 0.61, 0.88, 0.73, 0.34, 0.85, 0.98, 0.05, 0.74, 0.92, 0.5, 0.78, 0.26, 0.4, 0.27, 0.65, 0.71, 0.7, 0.8, 0.93, 0.36, 0.03, 0.45, 0.39, 0.77, 0.06]

Tier Two Merge:

[13B-airoboros-gpt4-1.4 + 13B-orca_mini_v2] offspring > [13B-gpt4-x-alpaca + 13B-Vicuna-cocktail] offspring

[0.2, 0.83, 0.24, 0.03, 0.37, 0.62, 0.02, 0.82, 0.65, 0.63, 0.45, 0.65, 0.48, 0.45, 0.24, 0.76, 0.06, 0.31, 0.45, 0.86, 0.23, 0.99, 0.93, 0.84, 0.96, 0.53, 0.95, 0.32, 0.19, 0.06, 0.4, 0.08, 0.62, 0.4, 0.26, 0.12, 0.16, 0.91, 0.14, 0.0]

Result:

13B-Ouroboros, a model that seems uncensored and highly competent. So far only Alpaca instruction prompting has been tested and seems to work solidly well.

Use:

Alpaca's instruct format can be used to do many things, including control of the terms of behavior between a user and a response from an agent in chat. Below is an example of a command injected into memory.

### Instruction:
Make Narrator function as a text based adventure game that responds with verbose, detailed, and creative descriptions of what happens next after Player's response.
Make Player function as the player input for Narrator's text based adventure game, controlling a character named (insert character name here, their short bio, and
whatever quest or other information to keep consistent in the interaction).

### Response:
{an empty new line here}

Language Models Used Credits:

13B-airoboros-gpt4-1.4 by jondurbin

https://huggingface.co/jondurbin/airoboros-13b-gpt4-1.4

13B-orca_mini_v2 by psmathur

https://huggingface.co/psmathur/orca_mini_v2_13b

13B-gpt4-x-alpaca by chavinlo

https://huggingface.co/chavinlo/gpt4-x-alpaca

13B-Vicuna-cocktail by reeducator

https://huggingface.co/reeducator/vicuna-13b-cocktail

Also thanks to Meta for LLaMA.

Each model was hand picked and considered for what it could contribute to this ensemble. Thanks to each and every one of you for your incredible work developing some of the best things to come out of this community.