<!-- 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-mod-datasets-txt-processing-rarity-all
This model is a fine-tuned version of gpt2 on the generator dataset. It achieves the following results on the evaluation set:
- Loss: 4.4313
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 |
---|---|---|---|
6.6762 | 0.3 | 500 | 5.6569 |
5.3312 | 0.6 | 1000 | 5.2263 |
4.9972 | 0.91 | 1500 | 4.9819 |
4.7192 | 1.21 | 2000 | 4.8383 |
4.5828 | 1.51 | 2500 | 4.7225 |
4.481 | 1.81 | 3000 | 4.6261 |
4.327 | 2.12 | 3500 | 4.5756 |
4.1937 | 2.42 | 4000 | 4.5233 |
4.1572 | 2.72 | 4500 | 4.4631 |
4.108 | 3.02 | 5000 | 4.4211 |
3.8775 | 3.33 | 5500 | 4.4180 |
3.8911 | 3.63 | 6000 | 4.3805 |
3.8636 | 3.93 | 6500 | 4.3392 |
3.6522 | 4.23 | 7000 | 4.3680 |
3.6068 | 4.54 | 7500 | 4.3577 |
3.6059 | 4.84 | 8000 | 4.3344 |
3.4722 | 5.14 | 8500 | 4.3582 |
3.3577 | 5.44 | 9000 | 4.3591 |
3.3543 | 5.75 | 9500 | 4.3534 |
3.3219 | 6.05 | 10000 | 4.3616 |
3.1978 | 6.35 | 10500 | 4.3687 |
3.2005 | 6.65 | 11000 | 4.3698 |
3.1998 | 6.96 | 11500 | 4.3697 |
Framework versions
- Transformers 4.26.1
- Pytorch 1.11.0+cu113
- Datasets 2.13.0
- Tokenizers 0.13.3