MiniCPM5-1B-Claude-Opus-Fable5-Thinking-AutoRound-W4A16-RTN

Model Details

This model is a int4 weight-only quantization with group_size 128 and symmetric quantization of GnLOLot/MiniCPM5-1B-Claude-Opus-Fable5-Thinking generated by AutoRound. Please follow the license of the original model.

Quantization Details

Attribute	Value
Base Model	GnLOLot/MiniCPM5-1B-Claude-Opus-Fable5-Thinking
Quantization Tool	AutoRound
Quantization Scheme	W4A16
Quantized Size	1102 MB

Evaluation Results

Task	Accuracy
hellaswag	0.3697
mmlu	0.4932
mmlu_abstract_algebra	0.3000
mmlu_anatomy	0.5704
mmlu_astronomy	0.5724
mmlu_business_ethics	0.5100
mmlu_clinical_knowledge	0.5509
mmlu_college_biology	0.4792
mmlu_college_chemistry	0.4000
mmlu_college_computer_science	0.4300
mmlu_college_mathematics	0.3400
mmlu_college_medicine	0.4566
mmlu_college_physics	0.3725
mmlu_computer_security	0.5800
mmlu_conceptual_physics	0.3915
mmlu_econometrics	0.3158
mmlu_electrical_engineering	0.5586
mmlu_elementary_mathematics	0.3783
mmlu_formal_logic	0.3333
mmlu_global_facts	0.3100
mmlu_high_school_biology	0.6129
mmlu_high_school_chemistry	0.4532
mmlu_high_school_computer_science	0.4400
mmlu_high_school_european_history	0.6061
mmlu_high_school_geography	0.6212
mmlu_high_school_government_and_politics	0.5751
mmlu_high_school_macroeconomics	0.4897
mmlu_high_school_mathematics	0.3370
mmlu_high_school_microeconomics	0.5504
mmlu_high_school_physics	0.3245
mmlu_high_school_psychology	0.6349
mmlu_high_school_statistics	0.3472
mmlu_high_school_us_history	0.5637
mmlu_high_school_world_history	0.6160
mmlu_human_aging	0.4888
mmlu_human_sexuality	0.6870
mmlu_humanities	0.4383
mmlu_international_law	0.7025
mmlu_jurisprudence	0.6296
mmlu_logical_fallacies	0.5460
mmlu_machine_learning	0.4286
mmlu_management	0.6408
mmlu_marketing	0.7009
mmlu_medical_genetics	0.6300
mmlu_miscellaneous	0.6909
mmlu_moral_disputes	0.5231
mmlu_moral_scenarios	0.2380
mmlu_nutrition	0.6176
mmlu_other	0.5629
mmlu_philosophy	0.5498
mmlu_prehistory	0.5370
mmlu_professional_accounting	0.3475
mmlu_professional_law	0.3625
mmlu_professional_medicine	0.4890
mmlu_professional_psychology	0.4984
mmlu_public_relations	0.5364
mmlu_security_studies	0.5551
mmlu_social_sciences	0.5635
mmlu_sociology	0.6816
mmlu_stem	0.4380
mmlu_us_foreign_policy	0.6900
mmlu_virology	0.4759
mmlu_world_religions	0.7135
piqa	0.6659

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-Claude-Opus-Fable5-Thinking-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-Claude-Opus-Fable5-Thinking-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.