This model is a int4 weight-only quantization with group_size 128 and symmetric quantization of nex-agi/Nex-N2-mini generated by TUNING. Please follow the license of the original model.

Quantization Details

Attribute	Value
Base Model	nex-agi/Nex-N2-mini
Quantization Tool	TUNING
Quantization Scheme	W4A16
Quantized Size	19504 MB

Evaluation Results

Task	Accuracy
hellaswag	0.6360
mmlu	0.8275
mmlu_abstract_algebra	0.6600
mmlu_anatomy	0.8667
mmlu_astronomy	0.9474
mmlu_business_ethics	0.8400
mmlu_clinical_knowledge	0.8906
mmlu_college_biology	0.9444
mmlu_college_chemistry	0.6400
mmlu_college_computer_science	0.7400
mmlu_college_mathematics	0.6600
mmlu_college_medicine	0.8150
mmlu_college_physics	0.7157
mmlu_computer_security	0.8300
mmlu_conceptual_physics	0.9362
mmlu_econometrics	0.7807
mmlu_electrical_engineering	0.8138
mmlu_elementary_mathematics	0.7804
mmlu_formal_logic	0.7063
mmlu_global_facts	0.5800
mmlu_high_school_biology	0.9484
mmlu_high_school_chemistry	0.8473
mmlu_high_school_computer_science	0.9300
mmlu_high_school_european_history	0.8727
mmlu_high_school_geography	0.9242
mmlu_high_school_government_and_politics	0.9741
mmlu_high_school_macroeconomics	0.8949
mmlu_high_school_mathematics	0.6296
mmlu_high_school_microeconomics	0.9664
mmlu_high_school_physics	0.7815
mmlu_high_school_psychology	0.9560
mmlu_high_school_statistics	0.8426
mmlu_high_school_us_history	0.9412
mmlu_high_school_world_history	0.9325
mmlu_human_aging	0.8206
mmlu_human_sexuality	0.9008
mmlu_humanities	0.7630
mmlu_international_law	0.9008
mmlu_jurisprudence	0.8889
mmlu_logical_fallacies	0.8896
mmlu_machine_learning	0.8571
mmlu_management	0.8932
mmlu_marketing	0.9487
mmlu_medical_genetics	0.9400
mmlu_miscellaneous	0.9323
mmlu_moral_disputes	0.8353
mmlu_moral_scenarios	0.6290
mmlu_nutrition	0.8856
mmlu_other	0.8590
mmlu_philosophy	0.8907
mmlu_prehistory	0.8920
mmlu_professional_accounting	0.7234
mmlu_professional_law	0.6662
mmlu_professional_medicine	0.9301
mmlu_professional_psychology	0.8905
mmlu_public_relations	0.7545
mmlu_security_studies	0.8041
mmlu_social_sciences	0.9035
mmlu_sociology	0.9204
mmlu_stem	0.8186
mmlu_us_foreign_policy	0.9200
mmlu_virology	0.6084
mmlu_world_religions	0.9006
piqa	0.8128

How to Use

HF Usage

Step 1: Install AutoRound

pip install auto-round

Step 2: Load and run the quantized model

python

from transformers import AutoModelForCausalLM, AutoTokenizer

model_name = "Nex-N2-mini-AutoRound-W4A16-Tuning"

# 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

bash

vllm serve Nex-N2-mini-AutoRound-W4A16-Tuning \
    --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.

LeaderboardModel1/Nex-N2-mini-AutoRound-W4A16-Tuning

Nex-N2-mini-AutoRound-W4A16-Tuning

Model Details