<!-- 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. -->
cbt-len
This model is a fine-tuned version of gpt2 on the generator dataset. It achieves the following results on the evaluation set:
- Loss: 4.0441
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 |
---|---|---|---|
6.3484 | 0.29 | 500 | 5.3167 |
5.0338 | 0.58 | 1000 | 4.8930 |
4.7064 | 0.87 | 1500 | 4.6570 |
4.4436 | 1.16 | 2000 | 4.5103 |
4.292 | 1.45 | 2500 | 4.3926 |
4.185 | 1.74 | 3000 | 4.2824 |
4.0816 | 2.03 | 3500 | 4.2049 |
3.8852 | 2.32 | 4000 | 4.1665 |
3.8585 | 2.61 | 4500 | 4.1129 |
3.8118 | 2.9 | 5000 | 4.0589 |
3.6437 | 3.18 | 5500 | 4.0555 |
3.5738 | 3.47 | 6000 | 4.0265 |
3.5595 | 3.76 | 6500 | 3.9952 |
3.4943 | 4.05 | 7000 | 3.9873 |
3.3042 | 4.34 | 7500 | 3.9849 |
3.2977 | 4.63 | 8000 | 3.9721 |
3.2891 | 4.92 | 8500 | 3.9592 |
3.1673 | 5.21 | 9000 | 3.9698 |
3.1192 | 5.5 | 9500 | 3.9685 |
3.1173 | 5.79 | 10000 | 3.9678 |
Framework versions
- Transformers 4.26.1
- Pytorch 1.11.0+cu113
- Datasets 2.13.0
- Tokenizers 0.13.3