Update README.md
Browse files
README.md
CHANGED
|
@@ -29,9 +29,8 @@ This was model was trained using LoRA, performing a random search over hyperpara
|
|
| 29 |
"target_modules": "Wqkv"
|
| 30 |
}
|
| 31 |
```
|
| 32 |
-
Here is a cleaner version:
|
| 33 |
|
| 34 |
-
The
|
| 35 |
## Test metrics
|
| 36 |
```json
|
| 37 |
{
|
|
|
|
| 29 |
"target_modules": "Wqkv"
|
| 30 |
}
|
| 31 |
```
|
|
|
|
| 32 |
|
| 33 |
+
The other hyperparameters used the Transformers Trainer defaults. Training used early stopping with a patience of 2; we report test set performance from the best checkpoint, selected by validation loss at epoch 8. The test Spearman's rho exceeds the average inter-annotator agreement, measured as each annotator's rho with the mean of all other annotators.
|
| 34 |
## Test metrics
|
| 35 |
```json
|
| 36 |
{
|