<!-- 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. -->
switchboard-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: 4.1008
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 |
---|---|---|---|
6.3582 | 0.29 | 500 | 5.3483 |
5.0337 | 0.58 | 1000 | 4.9322 |
4.7073 | 0.87 | 1500 | 4.6918 |
4.4439 | 1.17 | 2000 | 4.5506 |
4.294 | 1.46 | 2500 | 4.4317 |
4.187 | 1.75 | 3000 | 4.3272 |
4.0815 | 2.04 | 3500 | 4.2480 |
3.8891 | 2.33 | 4000 | 4.2093 |
3.8568 | 2.62 | 4500 | 4.1546 |
3.8319 | 2.92 | 5000 | 4.0999 |
3.6392 | 3.21 | 5500 | 4.0964 |
3.5919 | 3.5 | 6000 | 4.0644 |
3.5614 | 3.79 | 6500 | 4.0333 |
3.4752 | 4.08 | 7000 | 4.0305 |
3.3114 | 4.37 | 7500 | 4.0258 |
3.3071 | 4.66 | 8000 | 4.0137 |
3.2911 | 4.96 | 8500 | 3.9998 |
3.1578 | 5.25 | 9000 | 4.0124 |
3.1306 | 5.54 | 9500 | 4.0113 |
3.1228 | 5.83 | 10000 | 4.0107 |
Framework versions
- Transformers 4.26.1
- Pytorch 1.11.0+cu113
- Datasets 2.13.0
- Tokenizers 0.13.3