<!-- This model card has been generated automatically according to the information Keras had access to. You should probably proofread and complete it, then remove this comment. -->
distilgpt2-finetuned-eap
This model is a fine-tuned version of distilgpt2 on an unknown dataset. It achieves the following results on the evaluation set:
- Train Loss: 0.3557
- Train Accuracy: 0.0010
- Validation Loss: 7.4820
- Validation Accuracy: 0.0
- Epoch: 49
Model description
More information needed
Intended uses & limitations
More information needed
Training and evaluation data
More information needed
Training procedure
Training hyperparameters
The following hyperparameters were used during training:
- optimizer: {'name': 'AdamWeightDecay', 'learning_rate': 0.0002, 'decay': 0.0, 'beta_1': 0.9, 'beta_2': 0.999, 'epsilon': 1e-07, 'amsgrad': False, 'weight_decay_rate': 0.01}
- training_precision: float32
Training results
| Train Loss | Train Accuracy | Validation Loss | Validation Accuracy | Epoch |
|---|---|---|---|---|
| 4.6280 | 0.0015 | 4.3578 | 0.0013 | 0 |
| 4.2572 | 0.0018 | 4.2845 | 0.0013 | 1 |
| 4.0662 | 0.0018 | 4.2587 | 0.0012 | 2 |
| 3.9066 | 0.0016 | 4.2470 | 0.0012 | 3 |
| 3.7578 | 0.0016 | 4.2493 | 0.0013 | 4 |
| 3.6177 | 0.0018 | 4.2728 | 0.0013 | 5 |
| 3.4741 | 0.0017 | 4.3208 | 0.0012 | 6 |
| 3.3366 | 0.0017 | 4.3542 | 0.0012 | 7 |
| 3.1946 | 0.0016 | 4.3973 | 0.0012 | 8 |
| 3.0581 | 0.0016 | 4.4947 | 0.0014 | 9 |
| 2.9171 | 0.0017 | 4.5970 | 0.0013 | 10 |
| 2.7766 | 0.0016 | 4.6691 | 0.0014 | 11 |
| 2.6373 | 0.0015 | 4.7961 | 0.0012 | 12 |
| 2.4986 | 0.0014 | 4.8906 | 0.0002 | 13 |
| 2.3600 | 0.0015 | 4.9836 | 0.0002 | 14 |
| 2.2307 | 0.0014 | 5.1439 | 0.0001 | 15 |
| 2.1054 | 0.0015 | 5.3017 | 0.0001 | 16 |
| 1.9798 | 0.0015 | 5.4037 | 0.0001 | 17 |
| 1.8679 | 0.0014 | 5.5184 | 0.0001 | 18 |
| 1.7544 | 0.0013 | 5.6429 | 0.0001 | 19 |
| 1.6486 | 0.0013 | 5.7368 | 0.0001 | 20 |
| 1.5492 | 0.0013 | 5.8070 | 0.0001 | 21 |
| 1.4525 | 0.0013 | 5.9248 | 0.0001 | 22 |
| 1.3725 | 0.0013 | 5.9879 | 0.0 | 23 |
| 1.2901 | 0.0011 | 6.1063 | 0.0001 | 24 |
| 1.2178 | 0.0014 | 6.1828 | 0.0001 | 25 |
| 1.1429 | 0.0012 | 6.2581 | 0.0001 | 26 |
| 1.0831 | 0.0011 | 6.3003 | 0.0000 | 27 |
| 1.0266 | 0.0012 | 6.3558 | 0.0001 | 28 |
| 0.9673 | 0.0012 | 6.4831 | 0.0000 | 29 |
| 0.9116 | 0.0012 | 6.5555 | 0.0 | 30 |
| 0.8652 | 0.0012 | 6.6239 | 0.0 | 31 |
| 0.8198 | 0.0013 | 6.6751 | 0.0 | 32 |
| 0.7795 | 0.0013 | 6.7499 | 0.0 | 33 |
| 0.7410 | 0.0010 | 6.7741 | 0.0000 | 34 |
| 0.7015 | 0.0012 | 6.8395 | 0.0000 | 35 |
| 0.6679 | 0.0011 | 6.9150 | 0.0 | 36 |
| 0.6367 | 0.0010 | 6.9847 | 0.0 | 37 |
| 0.6034 | 0.0011 | 7.0334 | 0.0001 | 38 |
| 0.5756 | 0.0010 | 7.0516 | 0.0000 | 39 |
| 0.5442 | 0.0011 | 7.1220 | 0.0 | 40 |
| 0.5188 | 0.0010 | 7.1494 | 0.0000 | 41 |
| 0.4971 | 0.0010 | 7.2100 | 0.0 | 42 |
| 0.4711 | 0.0010 | 7.2883 | 0.0001 | 43 |
| 0.4501 | 0.0011 | 7.2946 | 0.0 | 44 |
| 0.4274 | 0.0011 | 7.3313 | 0.0001 | 45 |
| 0.4066 | 0.0011 | 7.3620 | 0.0000 | 46 |
| 0.3898 | 0.0011 | 7.4119 | 0.0000 | 47 |
| 0.3679 | 0.0010 | 7.4769 | 0.0001 | 48 |
| 0.3557 | 0.0010 | 7.4820 | 0.0 | 49 |
Framework versions
- Transformers 4.21.1
- TensorFlow 2.8.2
- Datasets 2.4.0
- Tokenizers 0.12.1
