<!-- 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-mod-no-repeating-sub-5p9k-length-5k
This model is a fine-tuned version of gpt2 on the generator dataset. It achieves the following results on the evaluation set:
- Loss: 3.1727
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 |
---|---|---|---|
6.7158 | 0.29 | 500 | 5.6284 |
5.343 | 0.59 | 1000 | 5.2028 |
4.9949 | 0.88 | 1500 | 4.9541 |
4.7239 | 1.18 | 2000 | 4.8156 |
4.5616 | 1.47 | 2500 | 4.6984 |
4.4605 | 1.76 | 3000 | 4.5956 |
4.3246 | 2.06 | 3500 | 4.5211 |
4.1396 | 2.35 | 4000 | 4.4691 |
4.1112 | 2.64 | 4500 | 4.4093 |
4.0706 | 2.94 | 5000 | 4.3647 |
3.8511 | 3.23 | 5500 | 4.3573 |
3.8089 | 3.53 | 6000 | 4.3333 |
3.7885 | 3.82 | 6500 | 4.2974 |
3.667 | 4.11 | 7000 | 4.2999 |
3.5215 | 4.41 | 7500 | 4.2987 |
3.5191 | 4.7 | 8000 | 4.2846 |
3.5029 | 4.99 | 8500 | 4.2727 |
3.328 | 5.29 | 9000 | 4.2904 |
3.3249 | 5.58 | 9500 | 4.2908 |
3.3222 | 5.88 | 10000 | 4.2899 |
Framework versions
- Transformers 4.26.1
- Pytorch 1.11.0+cu113
- Datasets 2.13.0
- Tokenizers 0.13.3