Update README.md
Browse files
README.md
CHANGED
|
@@ -3,6 +3,9 @@ Re-Upload of https://huggingface.co/dunzhang/stella_en_400M_v5 with :
|
|
| 3 |
- Max tokens lenght to 512 ( model has been trained on this sequence lenght )
|
| 4 |
- Padding strategy set to "BatchLongest"
|
| 5 |
|
|
|
|
|
|
|
| 6 |
Parameters at the end of the file "config.json" has been set manually to false for CPU usage:
|
| 7 |
```"unpad_inputs": false, "use_memory_efficient_attention": false```
|
|
|
|
| 8 |
You can turn them back to "true" to enable GPU back again
|
|
|
|
| 3 |
- Max tokens lenght to 512 ( model has been trained on this sequence lenght )
|
| 4 |
- Padding strategy set to "BatchLongest"
|
| 5 |
|
| 6 |
+
|
| 7 |
+
|
| 8 |
Parameters at the end of the file "config.json" has been set manually to false for CPU usage:
|
| 9 |
```"unpad_inputs": false, "use_memory_efficient_attention": false```
|
| 10 |
+
|
| 11 |
You can turn them back to "true" to enable GPU back again
|