Fara1.5-4B-AutoRound-W4A16-RTN

Model Details

This model is a int4 weight-only quantization with group_size 128 and symmetric quantization of microsoft/Fara1.5-4B generated by AutoRound. Please follow the license of the original model.

Quantization Details

Attribute	Value
Base Model	microsoft/Fara1.5-4B
Quantization Tool	AutoRound
Quantization Scheme	W4A16
Quantized Size	3619 MB

Evaluation Results

Task	Accuracy
hellaswag	0.5418
mmlu	0.7083
mmlu_abstract_algebra	0.4500
mmlu_anatomy	0.7259
mmlu_astronomy	0.8750
mmlu_business_ethics	0.8000
mmlu_clinical_knowledge	0.7811
mmlu_college_biology	0.8542
mmlu_college_chemistry	0.5500
mmlu_college_computer_science	0.6700
mmlu_college_mathematics	0.5200
mmlu_college_medicine	0.7052
mmlu_college_physics	0.4804
mmlu_computer_security	0.7800
mmlu_conceptual_physics	0.8000
mmlu_econometrics	0.6316
mmlu_electrical_engineering	0.7517
mmlu_elementary_mathematics	0.6481
mmlu_formal_logic	0.5556
mmlu_global_facts	0.3900
mmlu_high_school_biology	0.8935
mmlu_high_school_chemistry	0.7192
mmlu_high_school_computer_science	0.8200
mmlu_high_school_european_history	0.8061
mmlu_high_school_geography	0.8434
mmlu_high_school_government_and_politics	0.8912
mmlu_high_school_macroeconomics	0.7615
mmlu_high_school_mathematics	0.4889
mmlu_high_school_microeconomics	0.8613
mmlu_high_school_physics	0.6424
mmlu_high_school_psychology	0.8844
mmlu_high_school_statistics	0.7269
mmlu_high_school_us_history	0.8431
mmlu_high_school_world_history	0.8354
mmlu_human_aging	0.6816
mmlu_human_sexuality	0.7939
mmlu_humanities	0.6283
mmlu_international_law	0.8264
mmlu_jurisprudence	0.7593
mmlu_logical_fallacies	0.7669
mmlu_machine_learning	0.5893
mmlu_management	0.8835
mmlu_marketing	0.8932
mmlu_medical_genetics	0.8600
mmlu_miscellaneous	0.8072
mmlu_moral_disputes	0.6936
mmlu_moral_scenarios	0.4838
mmlu_nutrition	0.7909
mmlu_other	0.7470
mmlu_philosophy	0.7042
mmlu_prehistory	0.7932
mmlu_professional_accounting	0.5816
mmlu_professional_law	0.5124
mmlu_professional_medicine	0.7904
mmlu_professional_psychology	0.7353
mmlu_public_relations	0.6818
mmlu_security_studies	0.7510
mmlu_social_sciences	0.8027
mmlu_sociology	0.8657
mmlu_stem	0.6974
mmlu_us_foreign_policy	0.8800
mmlu_virology	0.4940
mmlu_world_religions	0.8246
piqa	0.7497

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 = "Fara1.5-4B-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 Fara1.5-4B-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.