<!-- 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-rarity-all-no-cut-rev
This model is a fine-tuned version of gpt2 on the generator dataset. It achieves the following results on the evaluation set:
- Loss: 4.3397
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 |
|---|---|---|---|
| 6.68 | 0.29 | 500 | 5.6382 |
| 5.326 | 0.59 | 1000 | 5.2100 |
| 4.9873 | 0.88 | 1500 | 4.9727 |
| 4.72 | 1.18 | 2000 | 4.8271 |
| 4.5629 | 1.47 | 2500 | 4.7084 |
| 4.468 | 1.76 | 3000 | 4.6087 |
| 4.3349 | 2.06 | 3500 | 4.5351 |
| 4.1534 | 2.35 | 4000 | 4.4838 |
| 4.1205 | 2.65 | 4500 | 4.4211 |
| 4.0865 | 2.94 | 5000 | 4.3663 |
| 3.8691 | 3.24 | 5500 | 4.3627 |
| 3.8207 | 3.53 | 6000 | 4.3272 |
| 3.8 | 3.82 | 6500 | 4.2943 |
| 3.6899 | 4.12 | 7000 | 4.2964 |
| 3.5382 | 4.41 | 7500 | 4.2861 |
| 3.5296 | 4.71 | 8000 | 4.2710 |
| 3.5189 | 5.0 | 8500 | 4.2564 |
| 3.3408 | 5.29 | 9000 | 4.2730 |
| 3.3436 | 5.59 | 9500 | 4.2712 |
| 3.3413 | 5.88 | 10000 | 4.2705 |
Framework versions
- Transformers 4.26.1
- Pytorch 1.11.0+cu113
- Datasets 2.13.0
- Tokenizers 0.13.3
