This model is a int4 weight-only quantization with group_size 128 and symmetric quantization of groxaxo/Qwevolution-V0-27B generated by AutoRound. Please follow the license of the original model.

Quantization Details

Attribute	Value
Base Model	groxaxo/Qwevolution-V0-27B
Quantization Tool	AutoRound
Quantization Scheme	W4A16
Quantized Size	17832 MB

Evaluation Results

Task	Accuracy
hellaswag	0.6718
mmlu	0.8536
mmlu_abstract_algebra	0.7900
mmlu_anatomy	0.8222
mmlu_astronomy	0.9671
mmlu_business_ethics	0.8100
mmlu_clinical_knowledge	0.9321
mmlu_college_biology	0.9722
mmlu_college_chemistry	0.6900
mmlu_college_computer_science	0.8400
mmlu_college_mathematics	0.7300
mmlu_college_medicine	0.8786
mmlu_college_physics	0.8137
mmlu_computer_security	0.8700
mmlu_conceptual_physics	0.9404
mmlu_econometrics	0.7719
mmlu_electrical_engineering	0.8345
mmlu_elementary_mathematics	0.8968
mmlu_formal_logic	0.7540
mmlu_global_facts	0.5500
mmlu_high_school_biology	0.9548
mmlu_high_school_chemistry	0.8522
mmlu_high_school_computer_science	0.9100
mmlu_high_school_european_history	0.8788
mmlu_high_school_geography	0.9293
mmlu_high_school_government_and_politics	0.9896
mmlu_high_school_macroeconomics	0.9333
mmlu_high_school_mathematics	0.6593
mmlu_high_school_microeconomics	0.9664
mmlu_high_school_physics	0.8609
mmlu_high_school_psychology	0.9523
mmlu_high_school_statistics	0.8704
mmlu_high_school_us_history	0.9461
mmlu_high_school_world_history	0.9409
mmlu_human_aging	0.8206
mmlu_human_sexuality	0.9084
mmlu_humanities	0.7979
mmlu_international_law	0.9091
mmlu_jurisprudence	0.8981
mmlu_logical_fallacies	0.8773
mmlu_machine_learning	0.7679
mmlu_management	0.9029
mmlu_marketing	0.9487
mmlu_medical_genetics	0.9700
mmlu_miscellaneous	0.9374
mmlu_moral_disputes	0.8497
mmlu_moral_scenarios	0.7385
mmlu_nutrition	0.9052
mmlu_other	0.8758
mmlu_philosophy	0.8714
mmlu_prehistory	0.9012
mmlu_professional_accounting	0.8050
mmlu_professional_law	0.7008
mmlu_professional_medicine	0.9522
mmlu_professional_psychology	0.8922
mmlu_public_relations	0.8000
mmlu_security_studies	0.8531
mmlu_social_sciences	0.9149
mmlu_sociology	0.9154
mmlu_stem	0.8551
mmlu_us_foreign_policy	0.9300
mmlu_virology	0.5663
mmlu_world_religions	0.9064
piqa	0.8058

How to Use

HF Usage

Step 1: Install AutoRound

pip install auto-round

Step 2: Load and run the quantized model

python

from transformers import AutoModelForCausalLM, AutoTokenizer

model_name = "Qwevolution-V0-27B-AutoRound-W4A16-RTN"

# load the tokenizer and the model
tokenizer = AutoTokenizer.from_pretrained(model_name)
model = AutoModelForCausalLM.from_pretrained(model_name, torch_dtype="auto", device_map="auto")

# prepare the model input
prompt = "Write a quick sort algorithm."
messages = [{"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)

# conduct text completion
generated_ids = model.generate(**model_inputs, max_new_tokens=512)
output_ids = generated_ids[0][len(model_inputs.input_ids[0]) :].tolist()

content = tokenizer.decode(output_ids, skip_special_tokens=True)
print("content:", content)

VLLM Usage

bash

vllm serve Qwevolution-V0-27B-AutoRound-W4A16-RTN \
    --trust-remote-code \
    --dtype bfloat16 \
    --tensor_parallel_size 1

If you encounter any issues, feel free to open an issue on the AutoRound GitHub repo or provide feedback on the Low-Bit Open LLM Leaderboard.

Ethical Considerations and Limitations

The model can produce factually incorrect output, and should not be relied on to produce factually accurate information. Because of the limitations of the pretrained model and the finetuning datasets, it is possible that this model could generate lewd, biased or otherwise offensive outputs. Therefore, before deploying any applications of the model, developers should perform safety testing.

Caveats and Recommendations

Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. Here are a couple of useful links to learn more about Intel's AI software:

Disclaimer

The license on this model does not constitute legal advice. We are not responsible for the actions of third parties who use this model. Please consult an attorney before using this model for commercial purposes.

Cite

@article{cheng2023optimize,
  title={Optimize weight rounding via signed gradient descent for the quantization of llms},
  author={Cheng, Wenhua and Zhang, Weiwei and Shen, Haihao and Cai, Yiyang and He, Xin and Lv, Kaokao and Liu, Yi},
  journal={arXiv preprint arXiv:2309.05516},
  year={2023}
}

arxiv github

This model is part of the Intel Low-Bit Open LLM Leaderboard initiative.

LeaderboardModel1/Qwevolution-V0-27B-AutoRound-W4A16-RTN

Qwevolution-V0-27B-AutoRound-W4A16-RTN

Model Details