Ornith-1.0-9B-Uncensored-AutoRound-W4A16-RTN

Model Details

This model is a int4 weight-only quantization with group_size 128 and symmetric quantization of PeppX/Ornith-1.0-9B-Uncensored generated by AutoRound. Please follow the license of the original model.

Quantization Details

Attribute	Value
Base Model	PeppX/Ornith-1.0-9B-Uncensored
Quantization Tool	AutoRound
Quantization Scheme	W4A16
Quantized Size	8180 MB

Evaluation Results

Task	Accuracy
hellaswag	0.5819
mmlu	0.7629
mmlu_abstract_algebra	0.6200
mmlu_anatomy	0.7704
mmlu_astronomy	0.9079
mmlu_business_ethics	0.8100
mmlu_clinical_knowledge	0.8453
mmlu_college_biology	0.9375
mmlu_college_chemistry	0.5800
mmlu_college_computer_science	0.6700
mmlu_college_mathematics	0.6000
mmlu_college_medicine	0.8035
mmlu_college_physics	0.6373
mmlu_computer_security	0.8200
mmlu_conceptual_physics	0.8894
mmlu_econometrics	0.7105
mmlu_electrical_engineering	0.8207
mmlu_elementary_mathematics	0.7751
mmlu_formal_logic	0.6111
mmlu_global_facts	0.5000
mmlu_high_school_biology	0.9387
mmlu_high_school_chemistry	0.7685
mmlu_high_school_computer_science	0.8600
mmlu_high_school_european_history	0.8606
mmlu_high_school_geography	0.9040
mmlu_high_school_government_and_politics	0.9637
mmlu_high_school_macroeconomics	0.8410
mmlu_high_school_mathematics	0.5185
mmlu_high_school_microeconomics	0.8992
mmlu_high_school_physics	0.6887
mmlu_high_school_psychology	0.9101
mmlu_high_school_statistics	0.7870
mmlu_high_school_us_history	0.8824
mmlu_high_school_world_history	0.9072
mmlu_human_aging	0.7937
mmlu_human_sexuality	0.8550
mmlu_humanities	0.6689
mmlu_international_law	0.8843
mmlu_jurisprudence	0.8148
mmlu_logical_fallacies	0.8405
mmlu_machine_learning	0.6518
mmlu_management	0.8738
mmlu_marketing	0.9316
mmlu_medical_genetics	0.8900
mmlu_miscellaneous	0.8825
mmlu_moral_disputes	0.7746
mmlu_moral_scenarios	0.4369
mmlu_nutrition	0.8529
mmlu_other	0.8114
mmlu_philosophy	0.7910
mmlu_prehistory	0.8179
mmlu_professional_accounting	0.6064
mmlu_professional_law	0.5782
mmlu_professional_medicine	0.8676
mmlu_professional_psychology	0.8186
mmlu_public_relations	0.7182
mmlu_security_studies	0.7551
mmlu_social_sciences	0.8554
mmlu_sociology	0.8955
mmlu_stem	0.7650
mmlu_us_foreign_policy	0.9100
mmlu_virology	0.5663
mmlu_world_religions	0.8421
piqa	0.7911

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 = "Ornith-1.0-9B-Uncensored-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 Ornith-1.0-9B-Uncensored-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.