Update README.md
Browse files
README.md
CHANGED
|
@@ -39,7 +39,7 @@ However, see an improvement, though not at the scale of DeepSeek's distilled mod
|
|
| 39 |
Non-commercial use.
|
| 40 |
|
| 41 |
## Training procedure
|
| 42 |
-
We used 8xH100 to train the model.
|
| 43 |
|
| 44 |
### Training hyperparameters
|
| 45 |
|
|
|
|
| 39 |
Non-commercial use.
|
| 40 |
|
| 41 |
## Training procedure
|
| 42 |
+
We used 8xH100 to train the model for 7 hours.
|
| 43 |
|
| 44 |
### Training hyperparameters
|
| 45 |
|