Update README.md
Browse files
README.md
CHANGED
|
@@ -47,7 +47,7 @@ Rather than training one model, we trained multiple models, each on a different
|
|
| 47 |
- Gradient Clipping: `1.0`
|
| 48 |
- Gradient Checkpointing: `true`
|
| 49 |
- Learning Rate: `2.5e-3`
|
| 50 |
-
- Eval Split: `0.
|
| 51 |
- Weight Decay: `0.01`
|
| 52 |
- Optimizer: `AdamW`
|
| 53 |
- AdamW Betas: `(0.9, 0.95)`
|
|
|
|
| 47 |
- Gradient Clipping: `1.0`
|
| 48 |
- Gradient Checkpointing: `true`
|
| 49 |
- Learning Rate: `2.5e-3`
|
| 50 |
+
- Eval Split: `0.00165`
|
| 51 |
- Weight Decay: `0.01`
|
| 52 |
- Optimizer: `AdamW`
|
| 53 |
- AdamW Betas: `(0.9, 0.95)`
|