<!-- 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. -->
gpt2-concat-cbt-rarity-2k-p3k
This model is a fine-tuned version of gpt2 on the generator dataset. It achieves the following results on the evaluation set:
- Loss: 3.0083
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: 7
- mixed_precision_training: Native AMP
Training results
Training Loss | Epoch | Step | Validation Loss |
---|---|---|---|
6.7186 | 0.29 | 500 | 5.6281 |
5.3685 | 0.58 | 1000 | 5.1947 |
5.0278 | 0.87 | 1500 | 4.9465 |
4.7459 | 1.17 | 2000 | 4.8014 |
4.5838 | 1.46 | 2500 | 4.6757 |
4.4777 | 1.75 | 3000 | 4.5664 |
4.3633 | 2.04 | 3500 | 4.4935 |
4.1601 | 2.33 | 4000 | 4.4512 |
4.1388 | 2.62 | 4500 | 4.3967 |
4.1004 | 2.91 | 5000 | 4.3434 |
3.9085 | 3.21 | 5500 | 4.3385 |
3.8559 | 3.5 | 6000 | 4.3100 |
3.8409 | 3.79 | 6500 | 4.2772 |
3.7507 | 4.08 | 7000 | 4.2758 |
3.5677 | 4.37 | 7500 | 4.2717 |
3.5771 | 4.66 | 8000 | 4.2566 |
3.5653 | 4.95 | 8500 | 4.2354 |
3.3565 | 5.24 | 9000 | 4.2632 |
3.3184 | 5.54 | 9500 | 4.2598 |
3.3222 | 5.83 | 10000 | 4.2510 |
3.2596 | 6.12 | 10500 | 4.2621 |
3.1718 | 6.41 | 11000 | 4.2643 |
3.1656 | 6.7 | 11500 | 4.2647 |
3.1666 | 6.99 | 12000 | 4.2645 |
Framework versions
- Transformers 4.26.1
- Pytorch 1.11.0+cu113
- Datasets 2.13.0
- Tokenizers 0.13.3