<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->
bnc_spoken-cbt-not-mixed-rarity-seed
This model is a fine-tuned version of gpt2 on the generator dataset. It achieves the following results on the evaluation set:
- Loss: 4.1204
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:
- learning_rate: 0.0005
- train_batch_size: 64
- eval_batch_size: 64
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_steps: 1000
- num_epochs: 6
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 6.3511 | 0.29 | 500 | 5.3480 |
| 5.0539 | 0.59 | 1000 | 4.9258 |
| 4.7241 | 0.88 | 1500 | 4.6921 |
| 4.4506 | 1.17 | 2000 | 4.5467 |
| 4.3095 | 1.46 | 2500 | 4.4347 |
| 4.2068 | 1.76 | 3000 | 4.3365 |
| 4.0909 | 2.05 | 3500 | 4.2622 |
| 3.9089 | 2.34 | 4000 | 4.2170 |
| 3.8733 | 2.63 | 4500 | 4.1643 |
| 3.8466 | 2.93 | 5000 | 4.1140 |
| 3.6507 | 3.22 | 5500 | 4.1107 |
| 3.5973 | 3.51 | 6000 | 4.0792 |
| 3.583 | 3.81 | 6500 | 4.0476 |
| 3.4873 | 4.1 | 7000 | 4.0505 |
| 3.3336 | 4.39 | 7500 | 4.0422 |
| 3.325 | 4.68 | 8000 | 4.0293 |
| 3.3172 | 4.98 | 8500 | 4.0176 |
| 3.1613 | 5.27 | 9000 | 4.0309 |
| 3.1512 | 5.56 | 9500 | 4.0298 |
| 3.1532 | 5.85 | 10000 | 4.0285 |
Framework versions
- Transformers 4.26.1
- Pytorch 1.11.0+cu113
- Datasets 2.13.0
- Tokenizers 0.13.3
