Commit
•
3fdc381
1
Parent(s):
85f7ce9
Update README.md
Browse files
README.md
CHANGED
|
@@ -1,6 +1,7 @@
|
|
| 1 |
** Converted model for GPTQ from https://huggingface.co/lmsys/vicuna-13b-delta-v0 **
|
| 2 |
|
| 3 |
* IMPORTANT NOTE: Use the .safetensors model unless it does not work. In which case, try .pt.
|
|
|
|
| 4 |
GPTQ conversion command (on CUDA branch):
|
| 5 |
CUDA_VISIBLE_DEVICES=0 python llama.py ../lmsys/vicuna-13b-v0 c4 --wbits 4 --true-sequential --groupsize 128 --save vicuna-13b-4bit-128g.pt
|
| 6 |
|
|
|
|
| 1 |
** Converted model for GPTQ from https://huggingface.co/lmsys/vicuna-13b-delta-v0 **
|
| 2 |
|
| 3 |
* IMPORTANT NOTE: Use the .safetensors model unless it does not work. In which case, try .pt.
|
| 4 |
+
|
| 5 |
GPTQ conversion command (on CUDA branch):
|
| 6 |
CUDA_VISIBLE_DEVICES=0 python llama.py ../lmsys/vicuna-13b-v0 c4 --wbits 4 --true-sequential --groupsize 128 --save vicuna-13b-4bit-128g.pt
|
| 7 |
|