<!-- 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-aochildes-16plus6k
This model is a fine-tuned version of gpt2 on the generator dataset. It achieves the following results on the evaluation set:
- Loss: 3.1978
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 |
---|---|---|---|
6.7265 | 0.3 | 500 | 5.6481 |
5.3801 | 0.59 | 1000 | 5.2065 |
5.0346 | 0.89 | 1500 | 4.9518 |
4.7589 | 1.19 | 2000 | 4.8123 |
4.6003 | 1.48 | 2500 | 4.6915 |
4.4941 | 1.78 | 3000 | 4.5806 |
4.3447 | 2.07 | 3500 | 4.5155 |
4.1761 | 2.37 | 4000 | 4.4640 |
4.1351 | 2.67 | 4500 | 4.4014 |
4.1043 | 2.96 | 5000 | 4.3576 |
3.8639 | 3.26 | 5500 | 4.3597 |
3.8432 | 3.56 | 6000 | 4.3266 |
3.8118 | 3.85 | 6500 | 4.2913 |
3.6736 | 4.15 | 7000 | 4.2957 |
3.5472 | 4.45 | 7500 | 4.2920 |
3.5398 | 4.74 | 8000 | 4.2794 |
3.507 | 5.04 | 8500 | 4.2806 |
3.3499 | 5.33 | 9000 | 4.2855 |
3.3504 | 5.63 | 9500 | 4.2851 |
3.3498 | 5.93 | 10000 | 4.2849 |
Framework versions
- Transformers 4.26.1
- Pytorch 1.11.0+cu113
- Datasets 2.13.0
- Tokenizers 0.13.3