SmolLM2-135M-AutoRound-W4A16-RTN

Model Details

This model is a int4 weight-only quantization with group_size 128 and symmetric quantization of HuggingFaceTB/SmolLM2-135M generated by AutoRound. Please follow the license of the original model.

Quantization Details

Attribute	Value
Base Model	HuggingFaceTB/SmolLM2-135M
Quantization Tool	AutoRound
Quantization Scheme	W4A16
Quantized Size	107 MB

Evaluation Results

Task	Accuracy
hellaswag	0.3413
mmlu	0.2316
mmlu_abstract_algebra	0.2300
mmlu_anatomy	0.1778
mmlu_astronomy	0.1974
mmlu_business_ethics	0.3000
mmlu_clinical_knowledge	0.2113
mmlu_college_biology	0.2778
mmlu_college_chemistry	0.2100
mmlu_college_computer_science	0.2700
mmlu_college_mathematics	0.1900
mmlu_college_medicine	0.2081
mmlu_college_physics	0.2059
mmlu_computer_security	0.2700
mmlu_conceptual_physics	0.2553
mmlu_econometrics	0.2544
mmlu_electrical_engineering	0.2483
mmlu_elementary_mathematics	0.2169
mmlu_formal_logic	0.2619
mmlu_global_facts	0.1800
mmlu_high_school_biology	0.1806
mmlu_high_school_chemistry	0.1921
mmlu_high_school_computer_science	0.2500
mmlu_high_school_european_history	0.2121
mmlu_high_school_geography	0.1717
mmlu_high_school_government_and_politics	0.2021
mmlu_high_school_macroeconomics	0.2000
mmlu_high_school_mathematics	0.2037
mmlu_high_school_microeconomics	0.2143
mmlu_high_school_physics	0.2053
mmlu_high_school_psychology	0.2000
mmlu_high_school_statistics	0.1481
mmlu_high_school_us_history	0.2598
mmlu_high_school_world_history	0.2911
mmlu_human_aging	0.3004
mmlu_human_sexuality	0.2595
mmlu_humanities	0.2425
mmlu_international_law	0.2314
mmlu_jurisprudence	0.2685
mmlu_logical_fallacies	0.2209
mmlu_machine_learning	0.3125
mmlu_management	0.1748
mmlu_marketing	0.2949
mmlu_medical_genetics	0.3000
mmlu_miscellaneous	0.2452
mmlu_moral_disputes	0.2514
mmlu_moral_scenarios	0.2380
mmlu_nutrition	0.2190
mmlu_other	0.2420
mmlu_philosophy	0.1865
mmlu_prehistory	0.2160
mmlu_professional_accounting	0.2518
mmlu_professional_law	0.2458
mmlu_professional_medicine	0.1912
mmlu_professional_psychology	0.2549
mmlu_public_relations	0.2182
mmlu_security_studies	0.1837
mmlu_social_sciences	0.2197
mmlu_sociology	0.2438
mmlu_stem	0.2166
mmlu_us_foreign_policy	0.2800
mmlu_virology	0.2771
mmlu_world_religions	0.3099
piqa	0.6578

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 = "SmolLM2-135M-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 SmolLM2-135M-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.