<!-- 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-guten-rarity-5k-2p5k
This model is a fine-tuned version of gpt2 on the generator dataset. It achieves the following results on the evaluation set:
- Loss: 3.1847
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 |
|---|---|---|---|
| 6.7001 | 0.3 | 500 | 5.6280 |
| 5.3666 | 0.59 | 1000 | 5.1990 |
| 5.0079 | 0.89 | 1500 | 4.9539 |
| 4.7385 | 1.19 | 2000 | 4.8095 |
| 4.5783 | 1.48 | 2500 | 4.6793 |
| 4.4688 | 1.78 | 3000 | 4.5716 |
| 4.3327 | 2.08 | 3500 | 4.4960 |
| 4.162 | 2.37 | 4000 | 4.4444 |
| 4.1218 | 2.67 | 4500 | 4.3820 |
| 4.0787 | 2.97 | 5000 | 4.3297 |
| 3.8425 | 3.26 | 5500 | 4.3301 |
| 3.825 | 3.56 | 6000 | 4.2940 |
| 3.8038 | 3.86 | 6500 | 4.2590 |
| 3.6546 | 4.15 | 7000 | 4.2647 |
| 3.5359 | 4.45 | 7500 | 4.2557 |
| 3.5282 | 4.75 | 8000 | 4.2377 |
| 3.4838 | 5.04 | 8500 | 4.2391 |
| 3.3383 | 5.34 | 9000 | 4.2426 |
| 3.3404 | 5.64 | 9500 | 4.2414 |
| 3.3337 | 5.93 | 10000 | 4.2410 |
Framework versions
- Transformers 4.26.1
- Pytorch 1.11.0+cu113
- Datasets 2.13.0
- Tokenizers 0.13.3
