Instructions to use vibhorag101/llama-2-7b-chat-hf-phr_mental_therapy_v2 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- PEFT
How to use vibhorag101/llama-2-7b-chat-hf-phr_mental_therapy_v2 with PEFT:
Task type is invalid.
- Notebooks
- Google Colab
- Kaggle
Update README.md
Browse files
README.md
CHANGED
|
@@ -45,6 +45,10 @@ The following hyperparameters were used during training:
|
|
| 45 |
- lr_scheduler_warmup_ratio: 0.03
|
| 46 |
- num_epochs: 1
|
| 47 |
- max_seq_length:1024
|
|
|
|
|
|
|
|
|
|
|
|
|
| 48 |
|
| 49 |
### Training results
|
| 50 |
|
|
|
|
| 45 |
- lr_scheduler_warmup_ratio: 0.03
|
| 46 |
- num_epochs: 1
|
| 47 |
- max_seq_length:1024
|
| 48 |
+
- Early Stopping:
|
| 49 |
+
- early_stopping_patience: 5
|
| 50 |
+
- early_stopping_threshold: 0.001
|
| 51 |
+
- parameter: eval_loss
|
| 52 |
|
| 53 |
### Training results
|
| 54 |
|