Darwin-28B-REASON-AutoRound-W4A16-RTN

Model Details

This model is a int4 weight-only quantization with group_size 128 and symmetric quantization of FINAL-Bench/Darwin-28B-REASON generated by AutoRound. Please follow the license of the original model.

Quantization Details

Attribute	Value
Base Model	FINAL-Bench/Darwin-28B-REASON
Quantization Tool	AutoRound
Quantization Scheme	W4A16
Quantized Size	17832 MB

Evaluation Results

Task	Accuracy
hellaswag	0.6489
mmlu	0.8512
mmlu_abstract_algebra	0.7000
mmlu_anatomy	0.8370
mmlu_astronomy	0.9408
mmlu_business_ethics	0.8300
mmlu_clinical_knowledge	0.9094
mmlu_college_biology	0.9722
mmlu_college_chemistry	0.6300
mmlu_college_computer_science	0.8000
mmlu_college_mathematics	0.7900
mmlu_college_medicine	0.8613
mmlu_college_physics	0.7353
mmlu_computer_security	0.8800
mmlu_conceptual_physics	0.9362
mmlu_econometrics	0.8509
mmlu_electrical_engineering	0.8552
mmlu_elementary_mathematics	0.8757
mmlu_formal_logic	0.7302
mmlu_global_facts	0.6000
mmlu_high_school_biology	0.9548
mmlu_high_school_chemistry	0.8374
mmlu_high_school_computer_science	0.9300
mmlu_high_school_european_history	0.9030
mmlu_high_school_geography	0.9495
mmlu_high_school_government_and_politics	0.9793
mmlu_high_school_macroeconomics	0.9205
mmlu_high_school_mathematics	0.7000
mmlu_high_school_microeconomics	0.9580
mmlu_high_school_physics	0.8013
mmlu_high_school_psychology	0.9468
mmlu_high_school_statistics	0.8519
mmlu_high_school_us_history	0.9461
mmlu_high_school_world_history	0.9620
mmlu_human_aging	0.8386
mmlu_human_sexuality	0.9008
mmlu_humanities	0.7989
mmlu_international_law	0.9256
mmlu_jurisprudence	0.9352
mmlu_logical_fallacies	0.9202
mmlu_machine_learning	0.7679
mmlu_management	0.9029
mmlu_marketing	0.9530
mmlu_medical_genetics	0.9600
mmlu_miscellaneous	0.9438
mmlu_moral_disputes	0.8497
mmlu_moral_scenarios	0.6693
mmlu_nutrition	0.9020
mmlu_other	0.8748
mmlu_philosophy	0.8810
mmlu_prehistory	0.9074
mmlu_professional_accounting	0.7908
mmlu_professional_law	0.7295
mmlu_professional_medicine	0.9265
mmlu_professional_psychology	0.8856
mmlu_public_relations	0.8000
mmlu_security_studies	0.8286
mmlu_social_sciences	0.9136
mmlu_sociology	0.9403
mmlu_stem	0.8452
mmlu_us_foreign_policy	0.9400
mmlu_virology	0.5783
mmlu_world_religions	0.9006
piqa	0.8134

How to Use

HF Usage

Step 1: Install AutoRound

pip install auto-round

Step 2: Load and run the quantized model

from transformers import AutoModelForCausalLM, AutoTokenizer

model_name = "Darwin-28B-REASON-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

vllm serve Darwin-28B-REASON-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:

• Intel Neural Compressor
• AutoRound

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.