OPEA
/

Safetensors
llama
4-bit precision
intel/auto-round
cicdatopea's picture
Update README.md
a51c6e2 verified
|
raw
history blame
8.16 kB
metadata
license: llama3.3
datasets:
  - NeelNanda/pile-10k

datasets: - NeelNanda/pile-10k

Model Details

This model is an int4 model with group_size 128 and symmetric quantization of meta-llama/Llama-3.3-70B-Instruct generated by intel/auto-round. Load the model with revision 12cbcc0 to use AutoGPTQ format

How To Use

Inference on CPU/HPU/CUDA

HPU: docker image with Gaudi Software Stack is recommended, please refer to following script for environment setup. More details can be found in Gaudi Guide.

from auto_round import AutoHfQuantizer ##must import for auto-round format
import torch
from transformers import AutoModelForCausalLM,AutoTokenizer
quantized_model_dir = "OPEA/Llama-3.3-70B-Instruct-int4-sym-inc"
tokenizer = AutoTokenizer.from_pretrained(quantized_model_dir)

model = AutoModelForCausalLM.from_pretrained(
    quantized_model_dir,
    torch_dtype='auto',
    device_map="auto",
    ##revision="12cbcc0", ##AutoGPTQ format
)

##import habana_frameworks.torch.core as htcore ## uncommnet it for HPU
##import habana_frameworks.torch.hpu as hthpu ## uncommnet it for HPU
##model = model.to(torch.bfloat16).to("hpu") ## uncommnet it for HPU

prompt = "There is a girl who likes adventure,"
messages = [
    {"role": "system", "content": "You are a helpful assistant."},
    {"role": "user", "content": prompt}
]

tokenizer = AutoTokenizer.from_pretrained(quantized_model_dir)
text = tokenizer.apply_chat_template(
    messages,
    tokenize=False,
    add_generation_prompt=True
)
model_inputs = tokenizer([text], return_tensors="pt").to(model.device)

generated_ids = model.generate(
    model_inputs.input_ids,
    max_new_tokens=200,  ##change this to align with the official usage
    do_sample=False  ##change this to align with the official usage
)
generated_ids = [
output_ids[len(input_ids):] for input_ids, output_ids in zip(model_inputs.input_ids, generated_ids)
]

response = tokenizer.batch_decode(generated_ids, skip_special_tokens=True)[0]
print(response)
##INT4:
## That sounds like the start of an exciting story. What kind of adventures does she like? Is she into hiking, traveling, trying new foods, or something else? Tell me more about her!

##BF16:
## That sounds like the start of an exciting story. The girl who likes adventure, let's call her Alex, is probably always looking for her next thrill. She might enjoy activities like hiking, rock climbing, or exploring new places. Perhaps she's always been drawn to the unknown and loves to challenge herself to try new things.

prompt = "Which one is larger, 9.11 or 9.8"
##INT4:
## 9.11 is larger than 9.8.

##BF16:
## 9.11 is larger than 9.8.

prompt = "How many r in strawberry."
##INT4:
## There are 2 R's in the word "strawberry".

##BF16:
## There are 2 R's in the word "strawberry".

prompt = "Once upon a time,"
##INT4:
## ...in a far-off kingdom, where the sun dipped into the horizon and painted the sky with hues of crimson and gold, there lived a young adventurer named Sophia. She was a curious and brave soul, with a heart full of wonder and a mind full of questions. Sophia lived in a small village on the outskirts of the kingdom, surrounded by rolling hills and dense forests that whispered secrets to the wind.
## One day, Sophia stumbled upon an ancient map that had been hidden away in the village library. The map was worn and torn, but it seemed to point to a mysterious location deep within the forest. The map was labeled with a single word: "Eldrador".
## Sophia felt an inexplicable pull towards the map and the secrets it held. She decided to embark on a journey to uncover the truth about Eldrador, and to explore the unknown lands that lay beyond the edge of the kingdom.
## As she set out on her quest, Sophia encountered a wise old wizard named Zephyr,

##BF16:
##  ...in a far-off kingdom, where the sun dipped into the horizon and painted the sky with hues of crimson and gold, there lived a young adventurer named Sophia. She had hair as black as the night and eyes as blue as the clearest summer sky. Sophia was known throughout the land for her bravery, kindness, and insatiable curiosity.
## What would you like to happen next in the story? Would you like Sophia to:
## A) Embark on a quest to find a legendary treasure
## B) Encounter a mysterious stranger with a hidden agenda
## C) Discover a magical forest filled with ancient secrets
## D) Something entirely different (please specify)
## Choose your response to progress the story!

Evaluate the model

pip3 install lm-eval==0.4.5

 auto-round --eval --model_name "OPEA/Llama-3.3-70B-Instruct-int4-sym-inc" --eval_bs 16  --tasks leaderboard_mmlu_pro,leaderboard_ifeval,lambada_openai,hellaswag,piqa,winogrande,truthfulqa_mc1,openbookqa,boolq,arc_easy,arc_challenge,mmlu,gsm8k
Metric BF16 INT4
avg 0.7023 0.7033
leaderboard_mmlu_pro 5shot 0.5484 0.5328
leaderboard_ifeval 0.6661=(0.7110+0.6211)/2 0.7132=(0.7554+0.6710)/2
mmlu 0.8195 0.8164
lambada_openai 0.7528 0.7599
hellaswag 0.6575 0.6540
winogrande 0.7869 0.7932
piqa 0.8303 0.8254
truthfulqa_mc1 0.4284 0.4272
openbookqa 0.3720 0.3540
boolq 0.8865 0.8826
arc_easy 0.8624 0.8577
arc_challenge 0.6109 0.6015
gsm8k(5shot) strict match 0.9083 0.9249

Generate the model

Here is the sample command to reproduce the model. We found auto-round is not stable for this model, please do NOT use --model_dtype "fp16"

auto-round  \
--model meta-llama/Llama-3.3-70B-Instruct \
--device 0 \
--group_size 128 \
--nsamples 512 \
--bits 4 \
--iter 1000 \
--disable_eval \
--low_gpu_mem_usage \
--format 'auto_gptq,auto_round' \
--output_dir "./tmp_autoround" 

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:

  • Intel Neural Compressor link

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