<!-- 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. -->
gpt2-3-dp-mod-aochild-cut
This model is a fine-tuned version of gpt2 on the generator dataset. It achieves the following results on the evaluation set:
- Loss: 4.4147
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: 10
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 6.7085 | 0.27 | 500 | 5.6381 |
| 5.3561 | 0.54 | 1000 | 5.2057 |
| 5.0129 | 0.81 | 1500 | 4.9548 |
| 4.7577 | 1.08 | 2000 | 4.8118 |
| 4.5699 | 1.35 | 2500 | 4.6948 |
| 4.4784 | 1.62 | 3000 | 4.5973 |
| 4.3982 | 1.89 | 3500 | 4.5120 |
| 4.2209 | 2.16 | 4000 | 4.4767 |
| 4.1372 | 2.42 | 4500 | 4.4233 |
| 4.1104 | 2.69 | 5000 | 4.3703 |
| 4.0829 | 2.96 | 5500 | 4.3281 |
| 3.8669 | 3.23 | 6000 | 4.3410 |
| 3.8631 | 3.5 | 6500 | 4.3058 |
| 3.8586 | 3.77 | 7000 | 4.2792 |
| 3.8067 | 4.04 | 7500 | 4.2780 |
| 3.5952 | 4.31 | 8000 | 4.2902 |
| 3.6276 | 4.58 | 8500 | 4.2712 |
| 3.6375 | 4.85 | 9000 | 4.2458 |
| 3.4886 | 5.12 | 9500 | 4.2834 |
| 3.3673 | 5.39 | 10000 | 4.2871 |
| 3.399 | 5.66 | 10500 | 4.2744 |
| 3.3996 | 5.93 | 11000 | 4.2531 |
| 3.1775 | 6.2 | 11500 | 4.3130 |
| 3.1317 | 6.47 | 12000 | 4.3148 |
| 3.1493 | 6.73 | 12500 | 4.3110 |
| 3.1562 | 7.0 | 13000 | 4.3110 |
| 2.8933 | 7.27 | 13500 | 4.3579 |
| 2.912 | 7.54 | 14000 | 4.3617 |
| 2.9199 | 7.81 | 14500 | 4.3607 |
| 2.8631 | 8.08 | 15000 | 4.3845 |
| 2.7354 | 8.35 | 15500 | 4.3967 |
| 2.7447 | 8.62 | 16000 | 4.4013 |
| 2.7529 | 8.89 | 16500 | 4.4021 |
| 2.6903 | 9.16 | 17000 | 4.4113 |
| 2.6512 | 9.43 | 17500 | 4.4138 |
| 2.6601 | 9.7 | 18000 | 4.4147 |
| 2.6617 | 9.97 | 18500 | 4.4147 |
Framework versions
- Transformers 4.26.1
- Pytorch 1.11.0+cu113
- Datasets 2.13.0
- Tokenizers 0.13.3
