File size: 5,027 Bytes
0c46da7
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
e885f32
0c46da7
 
 
 
 
e3673d3
0c46da7
c59e771
 
0c46da7
 
 
b3f236e
0c46da7
2b42c30
e3582e2
 
 
0c46da7
 
 
 
 
 
 
b3f236e
0c46da7
b3f236e
0c46da7
 
 
 
 
 
b23b349
0c46da7
 
 
 
 
 
 
 
b3f236e
0c46da7
 
c59e771
0c46da7
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
b23b349
8b50c3d
0c46da7
 
 
 
 
 
 
 
 
 
 
 
 
 
e885f32
0c46da7
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
---
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 `8.0` 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/abhinavkulkarni/llm-awq \
&& cd llm-awq \
&& git checkout ba01560f21516805fc5ceba5c2566dcbd1cf66d8 \
&& 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 snapshot_download

model_name = "abhinavkulkarni/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)
streamer = TextStreamer(tokenizer, skip_special_tokens=True)

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

load_quant = snapshot_download(model_name)

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,
    streamer=streamer)
```

## 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}
}
```