Model save
Browse files
README.md
CHANGED
|
@@ -15,8 +15,6 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 15 |
# sinhala_gpt2
|
| 16 |
|
| 17 |
This model is a fine-tuned version of [gpt2](https://huggingface.co/gpt2) on the None dataset.
|
| 18 |
-
It achieves the following results on the evaluation set:
|
| 19 |
-
- Loss: 0.6139
|
| 20 |
|
| 21 |
## Model description
|
| 22 |
|
|
@@ -44,20 +42,10 @@ The following hyperparameters were used during training:
|
|
| 44 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 45 |
- lr_scheduler_type: cosine
|
| 46 |
- lr_scheduler_warmup_steps: 500
|
| 47 |
-
- num_epochs:
|
| 48 |
|
| 49 |
### Training results
|
| 50 |
|
| 51 |
-
| Training Loss | Epoch | Step | Validation Loss |
|
| 52 |
-
|:-------------:|:-------:|:----:|:---------------:|
|
| 53 |
-
| 3.3922 | 7.3529 | 500 | 0.9294 |
|
| 54 |
-
| 0.862 | 14.7059 | 1000 | 0.7480 |
|
| 55 |
-
| 0.7413 | 22.0588 | 1500 | 0.6813 |
|
| 56 |
-
| 0.6795 | 29.4118 | 2000 | 0.6468 |
|
| 57 |
-
| 0.642 | 36.7647 | 2500 | 0.6274 |
|
| 58 |
-
| 0.6187 | 44.1176 | 3000 | 0.6192 |
|
| 59 |
-
| 0.607 | 51.4706 | 3500 | 0.6149 |
|
| 60 |
-
| 0.602 | 58.8235 | 4000 | 0.6139 |
|
| 61 |
|
| 62 |
|
| 63 |
### Framework versions
|
|
|
|
| 15 |
# sinhala_gpt2
|
| 16 |
|
| 17 |
This model is a fine-tuned version of [gpt2](https://huggingface.co/gpt2) on the None dataset.
|
|
|
|
|
|
|
| 18 |
|
| 19 |
## Model description
|
| 20 |
|
|
|
|
| 42 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 43 |
- lr_scheduler_type: cosine
|
| 44 |
- lr_scheduler_warmup_steps: 500
|
| 45 |
+
- num_epochs: 5
|
| 46 |
|
| 47 |
### Training results
|
| 48 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 49 |
|
| 50 |
|
| 51 |
### Framework versions
|