Commit
•
017af1c
1
Parent(s):
9502879
End of training
Browse files
README.md
CHANGED
|
@@ -12,6 +12,8 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 12 |
# llama-2-7b-chat-hf-ft-compact
|
| 13 |
|
| 14 |
This model was trained from scratch on an unknown dataset.
|
|
|
|
|
|
|
| 15 |
|
| 16 |
## Model description
|
| 17 |
|
|
@@ -42,6 +44,20 @@ The following hyperparameters were used during training:
|
|
| 42 |
- num_epochs: 3
|
| 43 |
- mixed_precision_training: Native AMP
|
| 44 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 45 |
### Framework versions
|
| 46 |
|
| 47 |
- Transformers 4.35.2
|
|
|
|
| 12 |
# llama-2-7b-chat-hf-ft-compact
|
| 13 |
|
| 14 |
This model was trained from scratch on an unknown dataset.
|
| 15 |
+
It achieves the following results on the evaluation set:
|
| 16 |
+
- Loss: 0.8585
|
| 17 |
|
| 18 |
## Model description
|
| 19 |
|
|
|
|
| 44 |
- num_epochs: 3
|
| 45 |
- mixed_precision_training: Native AMP
|
| 46 |
|
| 47 |
+
### Training results
|
| 48 |
+
|
| 49 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
| 50 |
+
|:-------------:|:-----:|:----:|:---------------:|
|
| 51 |
+
| 4.1287 | 0.36 | 25 | 2.8154 |
|
| 52 |
+
| 1.5 | 0.73 | 50 | 1.3773 |
|
| 53 |
+
| 1.1092 | 1.09 | 75 | 0.9817 |
|
| 54 |
+
| 0.9247 | 1.45 | 100 | 0.9045 |
|
| 55 |
+
| 0.8907 | 1.82 | 125 | 0.8791 |
|
| 56 |
+
| 0.8572 | 2.18 | 150 | 0.8663 |
|
| 57 |
+
| 0.8359 | 2.55 | 175 | 0.8608 |
|
| 58 |
+
| 0.8156 | 2.91 | 200 | 0.8585 |
|
| 59 |
+
|
| 60 |
+
|
| 61 |
### Framework versions
|
| 62 |
|
| 63 |
- Transformers 4.35.2
|