Commit
•
666fd67
1
Parent(s):
7e9f062
Model save
Browse files
README.md
CHANGED
|
@@ -12,8 +12,6 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 12 |
# llama-2-7b-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: 0.8405
|
| 17 |
|
| 18 |
## Model description
|
| 19 |
|
|
@@ -42,20 +40,7 @@ The following hyperparameters were used during training:
|
|
| 42 |
- lr_scheduler_type: linear
|
| 43 |
- lr_scheduler_warmup_ratio: 0.1
|
| 44 |
- num_epochs: 3
|
| 45 |
-
|
| 46 |
-
### Training results
|
| 47 |
-
|
| 48 |
-
| Training Loss | Epoch | Step | Validation Loss |
|
| 49 |
-
|:-------------:|:-----:|:----:|:---------------:|
|
| 50 |
-
| 3.9697 | 0.36 | 25 | 3.0220 |
|
| 51 |
-
| 1.1283 | 0.73 | 50 | 0.9941 |
|
| 52 |
-
| 0.9218 | 1.09 | 75 | 0.8991 |
|
| 53 |
-
| 0.8849 | 1.45 | 100 | 0.8700 |
|
| 54 |
-
| 0.8644 | 1.82 | 125 | 0.8561 |
|
| 55 |
-
| 0.8388 | 2.18 | 150 | 0.8465 |
|
| 56 |
-
| 0.8208 | 2.55 | 175 | 0.8417 |
|
| 57 |
-
| 0.8038 | 2.91 | 200 | 0.8405 |
|
| 58 |
-
|
| 59 |
|
| 60 |
### Framework versions
|
| 61 |
|
|
|
|
| 12 |
# llama-2-7b-hf-ft-compact
|
| 13 |
|
| 14 |
This model was trained from scratch on an unknown dataset.
|
|
|
|
|
|
|
| 15 |
|
| 16 |
## Model description
|
| 17 |
|
|
|
|
| 40 |
- lr_scheduler_type: linear
|
| 41 |
- lr_scheduler_warmup_ratio: 0.1
|
| 42 |
- num_epochs: 3
|
| 43 |
+
- mixed_precision_training: Native AMP
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 44 |
|
| 45 |
### Framework versions
|
| 46 |
|