Model Details: Mixtral-8x7B-Instruct-v0.1-int4-inc
This model is an int4 model with group_size 128 of mistralai/Mixtral-8x7B-Instruct-v0.1 generated by intel/auto-round. Layers "block_sparse_moe.gate" have not been quantized due to the exporting issue of AutoGPTQ format. Inference of this model is compatible with AutoGPTQ's Kernel.
How To Use
Reproduce the model
Here is the sample command to reproduce the model
git clone https://github.com/intel/auto-round
cd auto-round/examples/language-modeling
pip install -r requirements.txt
python3 main.py \
--model_name mistralai/Mixtral-8x7B-Instruct-v0.1 \
--device 0 \
--group_size 128 \
--bits 4 \
--iters 1000 \
--enable_minmax_tuning \
--nsamples 512 \
--deployment_device 'gpu' \
--scale_dtype 'fp32' \
--eval_bs 32 \
--output_dir "./tmp_autoround" \
--amp
Evaluate the model
Install lm-eval-harness from source, and the git id f3b7917091afba325af3980a35d8a6dcba03dc3f is used
lm_eval --model hf --model_args pretrained="Intel/Mixtral-8x7B-Instruct-v0.1-int4-inc",autogptq=True,gptq_use_triton=True --device cuda:0 --tasks lambada_openai,hellaswag,piqa,winogrande,truthfulqa_mc1,openbookqa,boolq,rte,arc_easy,arc_challenge,mmlu --batch_size 32
Metric | FP16 | INT4 |
---|---|---|
Avg. | 0.7000 | 0.6977 |
mmlu | 0.6885 | 0.6824 |
lambada_openai | 0.7718 | 0.7790 |
hellaswag | 0.6767 | 0.6745 |
winogrande | 0.7687 | 0.7719 |
piqa | 0.8351 | 0.8335 |
truthfulqa_mc1 | 0.4969 | 0.4884 |
openbookqa | 0.3680 | 0.3720 |
boolq | 0.8850 | 0.8783 |
rte | 0.7184 | 0.7004 |
arc_easy | 0.8699 | 0.8712 |
arc_challenge | 0.6220 | 0.6229 |
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.