Update README.md
Browse files
README.md
CHANGED
|
@@ -36,7 +36,7 @@ More information needed
|
|
| 36 |
|
| 37 |
The following hyperparameters were used during training:
|
| 38 |
- family_to_muon_lr = {
|
| 39 |
-
"language": _fallback(getattr(training_args, "language_muon_lr",
|
| 40 |
"vision": _fallback(getattr(training_args, "vision_muon_lr", 3e-5), vision_lr),
|
| 41 |
"merger": _fallback(getattr(training_args, "merger_muon_lr", 6e-5), merger_lr),
|
| 42 |
}
|
|
|
|
| 36 |
|
| 37 |
The following hyperparameters were used during training:
|
| 38 |
- family_to_muon_lr = {
|
| 39 |
+
"language": _fallback(getattr(training_args, "language_muon_lr", 2e-5), language_lr),
|
| 40 |
"vision": _fallback(getattr(training_args, "vision_muon_lr", 3e-5), vision_lr),
|
| 41 |
"merger": _fallback(getattr(training_args, "merger_muon_lr", 6e-5), merger_lr),
|
| 42 |
}
|