Update README.md
Browse files
README.md
CHANGED
|
@@ -107,5 +107,6 @@ I use the following configuration thanks to the suggestion of 金Goldkoron
|
|
| 107 |
- batch size 4
|
| 108 |
- conditional dropout 0.05
|
| 109 |
- change beta scheduler from `scaler_linear` to `linear` in `config.json` of the scheduler of the model
|
|
|
|
| 110 |
|
| 111 |
The released model is trained for 57751 steps, but among the provided checkpoints all the three starting from 34172 steps seem to work reasonably well.
|
|
|
|
| 107 |
- batch size 4
|
| 108 |
- conditional dropout 0.05
|
| 109 |
- change beta scheduler from `scaler_linear` to `linear` in `config.json` of the scheduler of the model
|
| 110 |
+
- clip skip 1
|
| 111 |
|
| 112 |
The released model is trained for 57751 steps, but among the provided checkpoints all the three starting from 34172 steps seem to work reasonably well.
|