README.md

---
license: cc
language:
- en
tags:
- AWQ
inference: false
---

# VMware/open-llama-7B-open-instruct (4-bit 128g AWQ Quantized)
[Instruction-tuned version](https://huggingface.co/VMware/open-llama-7b-open-instruct) of the fully trained [Open LLama 7B](https://huggingface.co/openlm-research/open_llama_7b) model.

This model is a 4-bit 128 group size AWQ quantized model. For more information about AWQ quantization, please click [here](https://github.com/mit-han-lab/llm-awq).

## Model Date

July 5, 2023

## Model License

Please refer to original OpenLLaMa model license ([link](https://huggingface.co/VMware/open-llama-7b-open-instruct)).

Please refer to the AWQ quantization license ([link](https://github.com/llm-awq/blob/main/LICENSE)).

## CUDA Version

This model was successfully tested on CUDA driver v530.30.02 and runtime v11.7 with Python v3.10.11. Please note that AWQ requires NVIDIA GPUs with compute capability of 80 or higher.

For Docker users, the `nvcr.io/nvidia/pytorch:23.06-py3` image is runtime v12.1 but otherwise the same as the configuration above and has also been verified to work.

## How to Use

```bash
git clone https://github.com/mit-han-lab/llm-awq \
&& cd llm-awq \
&& git checkout 71d8e68df78de6c0c817b029a568c064bf22132d \
&& pip install -e . \
&& cd awq/kernels \
&& python setup.py install
```

```python
import torch
from awq.quantize.quantizer import real_quantize_model_weight
from transformers import AutoModelForCausalLM, AutoConfig, AutoTokenizer
from accelerate import init_empty_weights, load_checkpoint_and_dispatch
from huggingface_hub import hf_hub_download

model_name = "VMware/open-llama-7b-open-instruct"

# Config
config = AutoConfig.from_pretrained(model_name, trust_remote_code=True)

# Tokenizer
tokenizer = AutoTokenizer.from_pretrained(config.tokenizer_name)

# Model
w_bit = 4
q_config = {
    "zero_point": True,
    "q_group_size": 128,
}

load_quant = hf_hub_download('abhinavkulkarni/VMWare-open-llama-7b-open-instruct-w4-g128-awq', 'pytorch_model.bin')

with init_empty_weights():
    model = AutoModelForCausalLM.from_config(config=config, 
                                                 torch_dtype=torch.float16, trust_remote_code=True)

real_quantize_model_weight(model, w_bit=w_bit, q_config=q_config, init_only=True)

model = load_checkpoint_and_dispatch(model, load_quant, device_map="balanced")

# Inference
prompt = f'''What is the difference between nuclear fusion and fission?
###Response:'''

input_ids = tokenizer(prompt, return_tensors='pt').input_ids.cuda()
output = model.generate(
    inputs=input_ids, 
    temperature=0.7,
    max_new_tokens=512,
    top_p=0.15,
    top_k=0,
    repetition_penalty=1.1,
    eos_token_id=tokenizer.eos_token_id
)
print(tokenizer.decode(output[0], skip_special_tokens=True))
```

## Evaluation

This evaluation was done using [LM-Eval](https://github.com/EleutherAI/lm-evaluation-harness).

[Open-LLaMA-7B-Instruct](https://huggingface.co/VMware/open-llama-7b-open-instruct)

|  Task  |Version|    Metric     | Value |   |Stderr|
|--------|------:|---------------|------:|---|------|
|wikitext|      1|word_perplexity|11.7531|   |      |
|        |       |byte_perplexity| 1.5853|   |      |
|        |       |bits_per_byte  | 0.6648|   |      |

[Open-LLaMA-7B-Instruct (4-bit 128-group AWQ)](https://huggingface.co/abhinavkulkarni/VMware-open-llama-7b-open-instruct-w4-g128-awq)

|  Task  |Version|    Metric     | Value |   |Stderr|
|--------|------:|---------------|------:|---|------|
|wikitext|      1|word_perplexity|12.1840|   |      |
|        |       |byte_perplexity| 1.5961|   |      |
|        |       |bits_per_byte  | 0.6745|   |      |

## Acknowledgements

If you found OpenLLaMA useful in your research or applications, please cite using the following BibTeX:
```
@software{openlm2023openllama,
  author = {Geng, Xinyang and Liu, Hao},
  title = {OpenLLaMA: An Open Reproduction of LLaMA},
  month = May,
  year = 2023,
  url = {https://github.com/openlm-research/open_llama}
}
```
```
@software{together2023redpajama,
  author = {Together Computer},
  title = {RedPajama-Data: An Open Source Recipe to Reproduce LLaMA training dataset},
  month = April,
  year = 2023,
  url = {https://github.com/togethercomputer/RedPajama-Data}
}
```
```
@article{touvron2023llama,
  title={Llama: Open and efficient foundation language models},
  author={Touvron, Hugo and Lavril, Thibaut and Izacard, Gautier and Martinet, Xavier and Lachaux, Marie-Anne and Lacroix, Timoth{\'e}e and Rozi{\`e}re, Baptiste and Goyal, Naman and Hambro, Eric and Azhar, Faisal and others},
  journal={arXiv preprint arXiv:2302.13971},
  year={2023}
}
```

The model was quantized with AWQ technique. If you find AWQ useful or relevant to your research, please kindly cite the paper:

```
@article{lin2023awq,
  title={AWQ: Activation-aware Weight Quantization for LLM Compression and Acceleration},
  author={Lin, Ji and Tang, Jiaming and Tang, Haotian and Yang, Shang and Dang, Xingyu and Han, Song},
  journal={arXiv},
  year={2023}
}
```