<!-- 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-mod-formatting-iorder-rarity-all-4k
This model is a fine-tuned version of gpt2 on the generator dataset. It achieves the following results on the evaluation set:
- Loss: 4.3158
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 |
---|---|---|---|
6.6962 | 0.29 | 500 | 5.6482 |
5.3352 | 0.59 | 1000 | 5.2168 |
4.9963 | 0.88 | 1500 | 4.9671 |
4.7147 | 1.17 | 2000 | 4.8164 |
4.5508 | 1.46 | 2500 | 4.6852 |
4.4503 | 1.76 | 3000 | 4.5766 |
4.3233 | 2.05 | 3500 | 4.4995 |
4.1239 | 2.34 | 4000 | 4.4513 |
4.0934 | 2.63 | 4500 | 4.3905 |
4.0645 | 2.93 | 5000 | 4.3376 |
3.8538 | 3.22 | 5500 | 4.3338 |
3.7937 | 3.51 | 6000 | 4.3034 |
3.781 | 3.8 | 6500 | 4.2718 |
3.6821 | 4.1 | 7000 | 4.2702 |
3.5082 | 4.39 | 7500 | 4.2633 |
3.5078 | 4.68 | 8000 | 4.2471 |
3.4936 | 4.97 | 8500 | 4.2346 |
3.34 | 5.27 | 9000 | 4.2492 |
3.3145 | 5.56 | 9500 | 4.2471 |
3.315 | 5.85 | 10000 | 4.2463 |
Framework versions
- Transformers 4.26.1
- Pytorch 1.11.0+cu113
- Datasets 2.13.0
- Tokenizers 0.13.3