Commit
•
c524c96
1
Parent(s):
b1ca4f7
Update README.md
Browse files
README.md
CHANGED
|
@@ -44,9 +44,7 @@ For example:
|
|
| 44 |
- lora_alpha = 16
|
| 45 |
- lora_dropout = 0
|
| 46 |
- bias = None
|
| 47 |
-
- use_gradient_checkpointing = "unsloth"
|
| 48 |
- random_state = 3407
|
| 49 |
-
Total Trainable Parameters: 41,943,040
|
| 50 |
|
| 51 |
* [ORPO Trainer](https://huggingface.co/docs/trl/main/en/orpo_trainer) Config
|
| 52 |
- num_epochs = 1
|
|
|
|
| 44 |
- lora_alpha = 16
|
| 45 |
- lora_dropout = 0
|
| 46 |
- bias = None
|
|
|
|
| 47 |
- random_state = 3407
|
|
|
|
| 48 |
|
| 49 |
* [ORPO Trainer](https://huggingface.co/docs/trl/main/en/orpo_trainer) Config
|
| 50 |
- num_epochs = 1
|