update model card README.md
Browse files
README.md
CHANGED
|
@@ -14,7 +14,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 14 |
|
| 15 |
This model is a fine-tuned version of [gpt2](https://huggingface.co/gpt2) on an unknown dataset.
|
| 16 |
It achieves the following results on the evaluation set:
|
| 17 |
-
- Loss: 2.
|
| 18 |
|
| 19 |
## Model description
|
| 20 |
|
|
@@ -39,41 +39,18 @@ The following hyperparameters were used during training:
|
|
| 39 |
- seed: 42
|
| 40 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 41 |
- lr_scheduler_type: linear
|
| 42 |
-
- lr_scheduler_warmup_steps:
|
| 43 |
- num_epochs: 3
|
| 44 |
|
| 45 |
### Training results
|
| 46 |
|
| 47 |
| Training Loss | Epoch | Step | Validation Loss |
|
| 48 |
|:-------------:|:-----:|:----:|:---------------:|
|
| 49 |
-
|
|
| 50 |
-
|
|
| 51 |
-
|
|
| 52 |
-
|
|
| 53 |
-
| 2.
|
| 54 |
-
| 2.9171 | 0.63 | 600 | 2.2268 |
|
| 55 |
-
| 2.9171 | 0.74 | 700 | 2.2035 |
|
| 56 |
-
| 2.9171 | 0.84 | 800 | 2.1836 |
|
| 57 |
-
| 2.9171 | 0.95 | 900 | 2.1626 |
|
| 58 |
-
| 2.5537 | 1.05 | 1000 | 2.1444 |
|
| 59 |
-
| 2.5537 | 1.16 | 1100 | 2.1314 |
|
| 60 |
-
| 2.5537 | 1.26 | 1200 | 2.1192 |
|
| 61 |
-
| 2.5537 | 1.37 | 1300 | 2.1096 |
|
| 62 |
-
| 2.5537 | 1.47 | 1400 | 2.0968 |
|
| 63 |
-
| 2.4399 | 1.58 | 1500 | 2.0868 |
|
| 64 |
-
| 2.4399 | 1.68 | 1600 | 2.0765 |
|
| 65 |
-
| 2.4399 | 1.79 | 1700 | 2.0675 |
|
| 66 |
-
| 2.4399 | 1.89 | 1800 | 2.0606 |
|
| 67 |
-
| 2.4399 | 2.0 | 1900 | 2.0556 |
|
| 68 |
-
| 2.4006 | 2.1 | 2000 | 2.0491 |
|
| 69 |
-
| 2.4006 | 2.21 | 2100 | 2.0454 |
|
| 70 |
-
| 2.4006 | 2.31 | 2200 | 2.0389 |
|
| 71 |
-
| 2.4006 | 2.42 | 2300 | 2.0372 |
|
| 72 |
-
| 2.4006 | 2.52 | 2400 | 2.0326 |
|
| 73 |
-
| 2.3532 | 2.63 | 2500 | 2.0316 |
|
| 74 |
-
| 2.3532 | 2.73 | 2600 | 2.0286 |
|
| 75 |
-
| 2.3532 | 2.84 | 2700 | 2.0276 |
|
| 76 |
-
| 2.3532 | 2.94 | 2800 | 2.0269 |
|
| 77 |
|
| 78 |
|
| 79 |
### Framework versions
|
|
|
|
| 14 |
|
| 15 |
This model is a fine-tuned version of [gpt2](https://huggingface.co/gpt2) on an unknown dataset.
|
| 16 |
It achieves the following results on the evaluation set:
|
| 17 |
+
- Loss: 2.0345
|
| 18 |
|
| 19 |
## Model description
|
| 20 |
|
|
|
|
| 39 |
- seed: 42
|
| 40 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 41 |
- lr_scheduler_type: linear
|
| 42 |
+
- lr_scheduler_warmup_steps: 500
|
| 43 |
- num_epochs: 3
|
| 44 |
|
| 45 |
### Training results
|
| 46 |
|
| 47 |
| Training Loss | Epoch | Step | Validation Loss |
|
| 48 |
|:-------------:|:-----:|:----:|:---------------:|
|
| 49 |
+
| 3.0017 | 0.53 | 500 | 2.3115 |
|
| 50 |
+
| 2.5821 | 1.05 | 1000 | 2.1626 |
|
| 51 |
+
| 2.4551 | 1.58 | 1500 | 2.0962 |
|
| 52 |
+
| 2.4095 | 2.1 | 2000 | 2.0540 |
|
| 53 |
+
| 2.3576 | 2.63 | 2500 | 2.0345 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 54 |
|
| 55 |
|
| 56 |
### Framework versions
|