Edit model card
TheBlokeAI

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


Anima Phi Neptune Mistral 7B - AWQ

Description

This repo contains AWQ model files for Severian's Anima Phi Neptune Mistral 7B.

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 Llama AWQ models for high-throughput concurrent inference in multi-user server scenarios.

As of September 25th 2023, preliminary Llama-only AWQ support has also been added to Huggingface Text Generation Inference (TGI).

Note that, at the time of writing, overall throughput is still lower than running vLLM or TGI 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.

Repositories available

Prompt template: INST

[INST] {prompt} [/INST]

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.

Branch Bits GS AWQ Dataset Seq Len Size
main 4 128 wikitext 4096 4.15 GB

Serving this model from vLLM

Documentation on installing and using vLLM can be found here.

Note: at the time of writing, vLLM has not yet done a new release with AWQ support.

If you try the vLLM examples below and get an error about quantization being unrecognised, or other AWQ-related issues, please install vLLM from Github source.

  • When using vLLM as a server, pass the --quantization awq parameter, for example:
python3 python -m vllm.entrypoints.api_server --model TheBloke/ANIMA-Phi-Neptune-Mistral-7B-AWQ --quantization awq --dtype half

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/ANIMA-Phi-Neptune-Mistral-7B-AWQ", quantization="awq", dtype="half")

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}")

Serving this model from Text Generation Inference (TGI)

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/ANIMA-Phi-Neptune-Mistral-7B-AWQ --port 3000 --quantize awq --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'''[INST] {prompt} [/INST]

'''

client = InferenceClient(endpoint_url)
response = client.text_generation(prompt,
                                  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}")

How to use this AWQ model from Python code

Install the necessary packages

Requires: AutoAWQ 0.1.1 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/ANIMA-Phi-Neptune-Mistral-7B-AWQ"

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

prompt = "Tell me about AI"
prompt_template=f'''[INST] {prompt} [/INST]

'''

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 should be possible with transformers pipeline as well in future
# But currently this is not yet supported by AutoAWQ (correct as of September 25th 2023)
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'])
"""

Compatibility

The files provided are tested to work with:

TGI merged AWQ support on September 25th, 2023: TGI PR #1054. Use the :latest Docker container until the next TGI release is made.

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.

Special thanks to: Aemon Algiz.

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

Thank you to all my generous patrons and donaters!

And thank you again to a16z for their generous grant.

Original model card: Severian's Anima Phi Neptune Mistral 7B

ANIMA-Phi-Neptune-Mistral-7B: Biomimicry Enhanced LLM

Overview

ANIMA (Advanced Nature Inspired Multidisciplinary Assistant) is an expert in various scientific disciplines, including but not limited to biomimicry, biology, and environmental science.


Model Description

ANIMA is fine-tuned on a rich dataset encompassing:

  • 4,000+ Nature-Biomimicry examples
  • 60k Biomimicry Design Process examples
  • 600k STEM facts from Wikipedia
  • Science/Philosophy focused 'All-You-Need-Is-Textbooks' dataset
  • Additional Tree of Knowledge + Biomimicry data combined fine-tuning

The model aims to assist users in solving problems using nature-inspired strategies and concepts.

Special Features

  • Multi-disciplinary Expertise: Knowledge across various scientific and philosophical domains.
  • Biomimicry Design Process: Incorporates a dataset generated by Mistral and Minotaur-15B. The dataset was then intricately processed by a real person to ensure factuality and grounding.

Datasets

  • Severian/Biomimicry (1st Fine-Tune)
  • emrgnt-cmplxty/sciphi-textbooks-are-all-you-need (2nd Fine-Tune)
  • Severian/Bio-Design-Process & fmars/wiki_stem (3rd Fine-Tune)
  • fblgit/tree-of-knowledge & Severian/Biomimicry` (4th Fine-Tune)
  • Combined all previous datasets into one giant dataset for a quick fine-tuning for a total of 730k data points (5th Fine-Tune)

Training Stages

  1. Base Model: ehartford/dolphin-2.0-mistral-7b
  2. 1st Fine-Tune: ANIMA-Echo-Mistral-7B-v1 - 1 Epoch x 5hrs on 1 x T4 Small
  3. 2nd Fine-Tune: ANIMA-Phi-Neptune-Mistral-v1-: 1 Epoch x 1hr on 1 x A100
  4. 3rd Fine-Tune: ANIMA-Phi-Neptune-Mistral-7B-v2 - 1 Epoch x 1hr on 1 x A100
  5. 4th Fine-Tune: 'ANIMA-Phi-Neptune-Mistral-7B-v3' - 1 Epoch x 1hr on 1 x A100
  6. 5th Fine-Tune: 'ANIMA-Phi-Neptune-Mistral-7B-v4' - 1 Epoch x 15mins on 1 x A100

πŸ“ Usage

ANIMA is designed around the Biomimicry Design Process to guide users in solving problems through biomimicry. For more information and examples, please visit Biomimicry Design Process.

🏁 Benchmarks

  • Average - 62.22
  • ARC - 56.83
  • HellaSwag - 78.82
  • MMLU - 53.84
  • TruthfulQA - 59.40

  • EXAMPLE OUTPUT

    [INST] How can biomimicry help in water purification? [/INST]

Biomimicry involves taking inspiration from natural processes and organisms to solve human problems. In the context of water purification, biomimicry can help develop sustainable and efficient methods for cleaning water. One specific example of a nature-inspired strategy to solve this problem is the use of biofiltration, which involves using microorganisms to remove pollutants from water.

Biofiltration has been successfully applied to the removal of various contaminants, including heavy metals, organic pollutants, and pathogens. For example, a study by Zhang et al. (2011) demonstrated that a mixed culture of Pseudomonas putida and Bacillus subtilis could effectively remove lead and cadmium from contaminated water. The researchers found that the biofilm formed by the mixed culture had a higher removal efficiency than that formed by either species alone. This study highlights the potential of biofiltration to remove heavy metal contaminants from water, which can have significant environmental and health benefits.

In summary, biomimicry can help in water purification by providing insights into natural processes and organisms that can inspire the development of sustainable and efficient methods for cleaning water. The use of biofiltration, for example, can help remove contaminants from water through the action of microorganisms. Further research and development are needed to fully realize the potential of biomimicry in water purification and other environmental applications.


πŸ“œ License

This project is licensed under MIT

Downloads last month
0
Safetensors
Model size
1.2B params
Tensor type
I32
Β·
FP16
Β·
Inference API
Inference API (serverless) has been turned off for this model.

Model tree for TheBloke/ANIMA-Phi-Neptune-Mistral-7B-AWQ

Adapter
(3)
this model

Datasets used to train TheBloke/ANIMA-Phi-Neptune-Mistral-7B-AWQ