<!-- 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-aochildes-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: 4.1424
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 |
---|---|---|---|
6.3658 | 0.29 | 500 | 5.3426 |
5.0625 | 0.59 | 1000 | 4.9354 |
4.7318 | 0.88 | 1500 | 4.7001 |
4.4693 | 1.18 | 2000 | 4.5584 |
4.3273 | 1.47 | 2500 | 4.4445 |
4.2267 | 1.77 | 3000 | 4.3439 |
4.0936 | 2.06 | 3500 | 4.2811 |
3.9235 | 2.36 | 4000 | 4.2297 |
3.9028 | 2.65 | 4500 | 4.1761 |
3.8511 | 2.95 | 5000 | 4.1320 |
3.6574 | 3.24 | 5500 | 4.1224 |
3.6199 | 3.54 | 6000 | 4.0962 |
3.5904 | 3.83 | 6500 | 4.0645 |
3.4759 | 4.12 | 7000 | 4.0696 |
3.346 | 4.42 | 7500 | 4.0637 |
3.337 | 4.71 | 8000 | 4.0533 |
3.3242 | 5.01 | 8500 | 4.0472 |
3.1638 | 5.3 | 9000 | 4.0575 |
3.1619 | 5.6 | 9500 | 4.0565 |
3.1557 | 5.89 | 10000 | 4.0557 |
Framework versions
- Transformers 4.26.1
- Pytorch 1.11.0+cu113
- Datasets 2.13.0
- Tokenizers 0.13.3