MiniCPM5-1B-AutoRound-W4A16-RTN

Model Details

This model is a int4 weight-only quantization with group_size 128 and symmetric quantization of openbmb/MiniCPM5-1B generated by AutoRound. Please follow the license of the original model.

Quantization Details

Attribute	Value
Base Model	openbmb/MiniCPM5-1B
Quantization Tool	AutoRound
Quantization Scheme	W4A16
Original Size	765 MB
Quantized Size	1102 MB

Evaluation Results

Task	Accuracy
hellaswag	0.3573
mmlu	0.4897
mmlu_abstract_algebra	0.2700
mmlu_anatomy	0.5481
mmlu_astronomy	0.5132
mmlu_business_ethics	0.4900
mmlu_clinical_knowledge	0.5547
mmlu_college_biology	0.5417
mmlu_college_chemistry	0.3900
mmlu_college_computer_science	0.3900
mmlu_college_mathematics	0.3500
mmlu_college_medicine	0.4855
mmlu_college_physics	0.3137
mmlu_computer_security	0.7000
mmlu_conceptual_physics	0.4468
mmlu_econometrics	0.2807
mmlu_electrical_engineering	0.5172
mmlu_elementary_mathematics	0.4074
mmlu_formal_logic	0.3651
mmlu_global_facts	0.3200
mmlu_high_school_biology	0.6419
mmlu_high_school_chemistry	0.4039
mmlu_high_school_computer_science	0.3800
mmlu_high_school_european_history	0.5879
mmlu_high_school_geography	0.5808
mmlu_high_school_government_and_politics	0.6010
mmlu_high_school_macroeconomics	0.4513
mmlu_high_school_mathematics	0.2926
mmlu_high_school_microeconomics	0.5378
mmlu_high_school_physics	0.2980
mmlu_high_school_psychology	0.6532
mmlu_high_school_statistics	0.3426
mmlu_high_school_us_history	0.5784
mmlu_high_school_world_history	0.5949
mmlu_human_aging	0.5022
mmlu_human_sexuality	0.6260
mmlu_humanities	0.4380
mmlu_international_law	0.6942
mmlu_jurisprudence	0.6296
mmlu_logical_fallacies	0.6319
mmlu_machine_learning	0.3750
mmlu_management	0.7087
mmlu_marketing	0.7222
mmlu_medical_genetics	0.5500
mmlu_miscellaneous	0.6794
mmlu_moral_disputes	0.4971
mmlu_moral_scenarios	0.2380
mmlu_nutrition	0.5882
mmlu_other	0.5636
mmlu_philosophy	0.5241
mmlu_prehistory	0.5494
mmlu_professional_accounting	0.3936
mmlu_professional_law	0.3670
mmlu_professional_medicine	0.4890
mmlu_professional_psychology	0.4804
mmlu_public_relations	0.5273
mmlu_security_studies	0.5714
mmlu_social_sciences	0.5525
mmlu_sociology	0.6716
mmlu_stem	0.4329
mmlu_us_foreign_policy	0.6800
mmlu_virology	0.4458
mmlu_world_religions	0.6725
piqa	0.6708

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 = "MiniCPM5-1B-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 MiniCPM5-1B-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.