cleanup empty lines, tweak env for runpod setup
Browse files- README.md +1 -0
- requirements.txt +0 -1
README.md
CHANGED
|
@@ -152,6 +152,7 @@ use_cpu: false
|
|
| 152 |
- I recommend `huggingface:transformers-pytorch-deepspeed-latest-gpu` see https://hub.docker.com/r/huggingface/transformers-pytorch-deepspeed-latest-gpu/
|
| 153 |
- Once you start your runpod, and SSH into it:
|
| 154 |
```shell
|
|
|
|
| 155 |
source <(curl -s https://raw.githubusercontent.com/winglian/axolotl/main/scripts/setup-runpod.sh)
|
| 156 |
```
|
| 157 |
|
|
|
|
| 152 |
- I recommend `huggingface:transformers-pytorch-deepspeed-latest-gpu` see https://hub.docker.com/r/huggingface/transformers-pytorch-deepspeed-latest-gpu/
|
| 153 |
- Once you start your runpod, and SSH into it:
|
| 154 |
```shell
|
| 155 |
+
export TORCH_CUDA_ARCH_LIST="7.0 7.5 8.0 8.6+PTX"
|
| 156 |
source <(curl -s https://raw.githubusercontent.com/winglian/axolotl/main/scripts/setup-runpod.sh)
|
| 157 |
```
|
| 158 |
|
requirements.txt
CHANGED
|
@@ -13,4 +13,3 @@ flash-attn
|
|
| 13 |
deepspeed
|
| 14 |
einops
|
| 15 |
xformers
|
| 16 |
-
|
|
|
|
| 13 |
deepspeed
|
| 14 |
einops
|
| 15 |
xformers
|
|
|