Update README.md
Browse files
README.md
CHANGED
@@ -9,11 +9,26 @@ https://github.com/qwopqwop200/GPTQ-for-LLaMa
|
|
9 |
|
10 |
LoRA credit to https://huggingface.co/baseten/alpaca-30b
|
11 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
12 |
# Usage
|
13 |
1. Run manually through GPTQ
|
14 |
2. (More setup but better UI) - Use the [text-generation-webui](https://github.com/oobabooga/text-generation-webui/wiki/LLaMA-model#4-bit-mode). Make sure to follow the installation steps first [here](https://github.com/oobabooga/text-generation-webui#installation) before adding GPTQ support.
|
15 |
|
16 |
-
**Note that a recent code change in GPTQ broke functionality for GPTQ in general, so please follow [these instructions](https://huggingface.co/elinas/alpaca-30b-lora-int4/discussions/2#641a38d5f1ad1c1173d8f192) to fix the issue!**
|
17 |
|
18 |
Since this is instruction tuned, for best results, use the following format for inference:
|
19 |
```
|
|
|
9 |
|
10 |
LoRA credit to https://huggingface.co/baseten/alpaca-30b
|
11 |
|
12 |
+
# Update 2023-03-27
|
13 |
+
New weights have been added. The old .pt version is no longer supported and has been replaced by a 128 groupsize safetensors file. Update to the latest GPTQ to use it.
|
14 |
+
|
15 |
+
**alpaca-30b-4bit-128g.safetensors**
|
16 |
+
|
17 |
+
Evals
|
18 |
+
-----
|
19 |
+
**c4-new** -
|
20 |
+
coming soon
|
21 |
+
|
22 |
+
**ptb-new** -
|
23 |
+
coming soon
|
24 |
+
|
25 |
+
**wikitext2** -
|
26 |
+
coming soon
|
27 |
+
|
28 |
# Usage
|
29 |
1. Run manually through GPTQ
|
30 |
2. (More setup but better UI) - Use the [text-generation-webui](https://github.com/oobabooga/text-generation-webui/wiki/LLaMA-model#4-bit-mode). Make sure to follow the installation steps first [here](https://github.com/oobabooga/text-generation-webui#installation) before adding GPTQ support.
|
31 |
|
|
|
32 |
|
33 |
Since this is instruction tuned, for best results, use the following format for inference:
|
34 |
```
|