File size: 2,412 Bytes
02a3f91
 
277d084
 
02a3f91
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
277d084
 
 
 
 
 
 
 
 
02a3f91
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
cdc960d
 
 
 
 
 
 
 
 
 
 
 
02a3f91
cdc960d
 
02a3f91
cdc960d
 
02a3f91
 
 
 
 
 
 
 
 
 
 
 
 
 
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
---
base_model: bigscience/bloom-3b
datasets:
- c4
language:
- ak
- ar
- as
- bm
- bn
- ca
- code
- en
- es
- eu
- fon
- fr
- gu
- hi
- id
- ig
- ki
- kn
- lg
- ln
- ml
- mr
- ne
- nso
- ny
- or
- pa
- pt
- rn
- rw
- sn
- st
- sw
- ta
- te
- tn
- ts
- tum
- tw
- ur
- vi
- wo
- xh
- yo
- zh
- zhs
- zht
- zu
license: bigscience-bloom-rail-1.0
model_name: bloom-3b
pipeline_tag: text-generation
tags:
- pretrained
inference: false
model_creator: bigscience
model_type: bloom
quantized_by: iproskurina
---



# 🌸 BLOOM 3B - GPTQ
- Model creator: [BigScience](https://huggingface.co/bigscience)
- Original model: [BLOOM 3B](https://huggingface.co/bigscience/bloom-3b)

The model published in this repo was quantized to 4bit using [AutoGPTQ](https://github.com/PanQiWei/AutoGPTQ).

**Quantization details**

**All quantization parameters were taken from [GPTQ paper](https://arxiv.org/abs/2210.17323).**

GPTQ calibration data consisted of 128 random 2048 token segments from the [C4 dataset](https://huggingface.co/datasets/c4).

The grouping size used for quantization is equal to 128.

## How to use this GPTQ model from Python code

### Install the necessary packages

Requires: Transformers 4.33.0 or later, Optimum 1.12.0 or later, and AutoGPTQ 0.4.2 or later.

```shell
pip3 install --upgrade transformers optimum
# If using PyTorch 2.1 + CUDA 12.x:
pip3 install --upgrade auto-gptq
# or, if using PyTorch 2.1 + CUDA 11.x:
pip3 install --upgrade auto-gptq --extra-index-url https://huggingface.github.io/autogptq-index/whl/cu118/
```

If you are using PyTorch 2.0, you will need to install AutoGPTQ from source. Likewise if you have problems with the pre-built wheels, you should try building from source:

```shell
pip3 uninstall -y auto-gptq
git clone https://github.com/PanQiWei/AutoGPTQ
cd AutoGPTQ
git checkout v0.5.1
pip3 install .
```

### You can then use the following code

```python

from transformers import AutoTokenizer, TextGenerationPipeline,AutoModelForCausalLM
from auto_gptq import AutoGPTQForCausalLM, BaseQuantizeConfig
pretrained_model_dir = "iproskurina/bloom-3b-gptq-4bit"
tokenizer = AutoTokenizer.from_pretrained(pretrained_model_dir, use_fast=True)
model = AutoGPTQForCausalLM.from_quantized(pretrained_model_dir, device="cuda:0", model_basename="model")
pipeline = TextGenerationPipeline(model=model, tokenizer=tokenizer)
print(pipeline("auto-gptq is")[0]["generated_text"])
```