Commit
•
dc68718
1
Parent(s):
0ab711f
Create model card
Browse files
README.md
ADDED
@@ -0,0 +1,29 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
library_name: transformers
|
3 |
+
tags:
|
4 |
+
- llama
|
5 |
+
- facebook
|
6 |
+
- meta
|
7 |
+
- llama-2
|
8 |
+
- conversational
|
9 |
+
- text-generation-inference
|
10 |
+
---
|
11 |
+
|
12 |
+
A quantization of [meta-llama/Llama-2-7b](https://huggingface.co/meta-llama/Llama-2-7b) using [PV-Tuning](https://arxiv.org/abs/2405.14852) on top of [AQLM](https://arxiv.org/abs/2401.06118) quantization
|
13 |
+
|
14 |
+
|
15 |
+
For this quantization, we used 1 codebook of 16 bits for groups of 8 weights.
|
16 |
+
|
17 |
+
|
18 |
+
| Model | AQLM scheme | WikiText 2 PPL | Model size, Gb | Hub link |
|
19 |
+
|------------|-------------|----------------|----------------|--------------------------------------------------------------------------|
|
20 |
+
| Llama-2-7b (this) | 1x16g8 | 5.68 | 2.4 | [Link](https://huggingface.co/ISTA-DASLab/Llama-2-7b-AQLM-PV-2Bit-1x16-hf) |
|
21 |
+
| Llama-2-7b | 2x8g8 | 5.90 | 2.2 | [Link](https://huggingface.co/ISTA-DASLab/Llama-2-7b-AQLM-PV-2Bit-2x8-hf) |
|
22 |
+
| Llama-2-13b| 1x16g8 | 5.05 | 4.1 | [Link](https://huggingface.co/ISTA-DASLab/Llama-2-13b-AQLM-PV-2Bit-1x16-hf)|
|
23 |
+
| Llama-2-70b| 1x16g8 | 3.78 | 18.8 | [Link](https://huggingface.co/ISTA-DASLab/Llama-2-70b-AQLM-PV-2Bit-1x16-hf)|
|
24 |
+
|
25 |
+
The 1x16g16 (1-bit) models are on the way, as soon as we update the inference lib with their respective kernels.
|
26 |
+
|
27 |
+
To learn more about the inference, as well as the information on how to quantize models yourself, please refer to the [official GitHub repo](https://github.com/Vahe1994/AQLM).
|
28 |
+
The original code for PV-Tuning can be found in the [AQLM@pv-tuning](https://github.com/Vahe1994/AQLM/tree/pv-tuning) branch.
|
29 |
+
|