<!-- 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-log-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: 4.1333
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 |
---|---|---|---|
6.3566 | 0.29 | 500 | 5.3408 |
5.0449 | 0.58 | 1000 | 4.9324 |
4.7171 | 0.88 | 1500 | 4.6973 |
4.4481 | 1.17 | 2000 | 4.5580 |
4.31 | 1.46 | 2500 | 4.4393 |
4.2064 | 1.75 | 3000 | 4.3418 |
4.0828 | 2.05 | 3500 | 4.2678 |
3.8965 | 2.34 | 4000 | 4.2172 |
3.8712 | 2.63 | 4500 | 4.1668 |
3.8382 | 2.92 | 5000 | 4.1164 |
3.6437 | 3.22 | 5500 | 4.1146 |
3.5911 | 3.51 | 6000 | 4.0877 |
3.5736 | 3.8 | 6500 | 4.0548 |
3.4773 | 4.09 | 7000 | 4.0578 |
3.3209 | 4.39 | 7500 | 4.0494 |
3.3194 | 4.68 | 8000 | 4.0397 |
3.3069 | 4.97 | 8500 | 4.0284 |
3.1506 | 5.26 | 9000 | 4.0416 |
3.1422 | 5.56 | 9500 | 4.0408 |
3.138 | 5.85 | 10000 | 4.0400 |
Framework versions
- Transformers 4.26.1
- Pytorch 1.11.0+cu113
- Datasets 2.13.0
- Tokenizers 0.13.3