osmQwopus3.6-27B-Fable-Agentic-AutoRound-W4A16-RTN

Model Details

This model is a int4 weight-only quantization with group_size 128 and symmetric quantization of osmapi/osmQwopus3.6-27B-Fable-Agentic generated by AutoRound. Please follow the license of the original model.

Quantization Details

Attribute	Value
Base Model	osmapi/osmQwopus3.6-27B-Fable-Agentic
Quantization Tool	AutoRound
Quantization Scheme	W4A16
Quantized Size	17832 MB

Evaluation Results

Task	Accuracy
hellaswag	0.6475
mmlu	0.8541
mmlu_abstract_algebra	0.7800
mmlu_anatomy	0.8296
mmlu_astronomy	0.9276
mmlu_business_ethics	0.8300
mmlu_clinical_knowledge	0.9132
mmlu_college_biology	0.9514
mmlu_college_chemistry	0.6500
mmlu_college_computer_science	0.8500
mmlu_college_mathematics	0.7600
mmlu_college_medicine	0.8728
mmlu_college_physics	0.7255
mmlu_computer_security	0.8800
mmlu_conceptual_physics	0.9489
mmlu_econometrics	0.7895
mmlu_electrical_engineering	0.8345
mmlu_elementary_mathematics	0.8704
mmlu_formal_logic	0.7619
mmlu_global_facts	0.5600
mmlu_high_school_biology	0.9613
mmlu_high_school_chemistry	0.8473
mmlu_high_school_computer_science	0.9200
mmlu_high_school_european_history	0.9030
mmlu_high_school_geography	0.9293
mmlu_high_school_government_and_politics	0.9845
mmlu_high_school_macroeconomics	0.9333
mmlu_high_school_mathematics	0.6407
mmlu_high_school_microeconomics	0.9622
mmlu_high_school_physics	0.8079
mmlu_high_school_psychology	0.9541
mmlu_high_school_statistics	0.8519
mmlu_high_school_us_history	0.9510
mmlu_high_school_world_history	0.9536
mmlu_human_aging	0.8475
mmlu_human_sexuality	0.9313
mmlu_humanities	0.8074
mmlu_international_law	0.9421
mmlu_jurisprudence	0.9259
mmlu_logical_fallacies	0.9141
mmlu_machine_learning	0.8036
mmlu_management	0.9029
mmlu_marketing	0.9658
mmlu_medical_genetics	0.9400
mmlu_miscellaneous	0.9425
mmlu_moral_disputes	0.8237
mmlu_moral_scenarios	0.7385
mmlu_nutrition	0.9183
mmlu_other	0.8761
mmlu_philosophy	0.8810
mmlu_prehistory	0.9043
mmlu_professional_accounting	0.7837
mmlu_professional_law	0.7197
mmlu_professional_medicine	0.9522
mmlu_professional_psychology	0.8856
mmlu_public_relations	0.8182
mmlu_security_studies	0.8286
mmlu_social_sciences	0.9139
mmlu_sociology	0.9204
mmlu_stem	0.8436
mmlu_us_foreign_policy	0.9300
mmlu_virology	0.5361
mmlu_world_religions	0.9006
piqa	0.8215

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 = "osmQwopus3.6-27B-Fable-Agentic-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 osmQwopus3.6-27B-Fable-Agentic-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.