license: apache-2.0
datasets:
- NeelNanda/pile-10k
Model Details: Mistral-7B-v0.1-int4-inc-lmhead
This model is an int4 model with group_size 128 and quantized lmhead of mistralai/Mistral-7B-v0.1 generated by intel/auto-round.
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/Mistral-7B-v0.1 \
--device 0 \
--group_size 128 \
--bits 4 \
--iters 1000 \
--quant_lm_head \
--disable_low_gpu_mem_usage \
--deployment_device 'gpu' \
--output_dir "./tmp_autoround"
Use the model
pip install auto-gptq
Install auto-round from source first
from transformers import AutoModelForCausalLM, AutoTokenizer
from auto_round.auto_quantizer import AutoHfQuantizer
quantized_model_dir = "Intel/Mistral-7B-v0.1-int4-inc-lmhead"
model = AutoModelForCausalLM.from_pretrained(quantized_model_dir,
device_map="auto",
trust_remote_code=False,
)
tokenizer = AutoTokenizer.from_pretrained(quantized_model_dir, use_fast=True)
print(tokenizer.decode(model.generate(**tokenizer("There is a girl who likes adventure,", return_tensors="pt").to(model.device),max_new_tokens=50)[0]))
Evaluate the model
pip install lm-eval==0.4.2
git clone https://github.com/intel/auto-round
cd auto-round/examples/language-modeling
python3 eval_042/evluation.py --model_name "Intel/Mistral-7B-v0.1-int4-inc-lmhead" --eval_bs 32
| Metric | BF16 | INT4-lmhead | INT4 |
|---|---|---|---|
| Avg. | 0.6260 | 0.6228 | 0.6218 |
| mmlu | 0.5868 | 0.5760 | 0.5772 |
| lambada_openai | 0.7555 | 0.7539 | 0.7543 |
| hellaswag | 0.6125 | 0.6055 | 0.6072 |
| winogrande | 0.7395 | 0.7380 | 0.7388 |
| piqa | 0.8069 | 0.8009 | 0.8030 |
| truthfulqa_mc1 | 0.2803 | 0.2876 | 0.2864 |
| openbookqa | 0.3280 | 0.3300 | 0.3260 |
| boolq | 0.8379 | 0.8291 | 0.8281 |
| arc_easy | 0.8089 | 0.8043 | 0.8035 |
| arc_challenge | 0.5034 | 0.5026 | 0.4932 |
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.