<!-- 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-len-no-cut
This model is a fine-tuned version of gpt2 on the generator dataset. It achieves the following results on the evaluation set:
- Loss: 4.3508
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 |
---|---|---|---|
6.7243 | 0.29 | 500 | 5.6347 |
5.3457 | 0.59 | 1000 | 5.2001 |
4.9946 | 0.88 | 1500 | 4.9636 |
4.7281 | 1.17 | 2000 | 4.8150 |
4.558 | 1.47 | 2500 | 4.6913 |
4.4561 | 1.76 | 3000 | 4.5889 |
4.3262 | 2.05 | 3500 | 4.5179 |
4.1342 | 2.34 | 4000 | 4.4699 |
4.1059 | 2.64 | 4500 | 4.4095 |
4.0656 | 2.93 | 5000 | 4.3560 |
3.8596 | 3.22 | 5500 | 4.3558 |
3.8032 | 3.52 | 6000 | 4.3312 |
3.7915 | 3.81 | 6500 | 4.2977 |
3.6736 | 4.1 | 7000 | 4.3046 |
3.5203 | 4.4 | 7500 | 4.2990 |
3.5147 | 4.69 | 8000 | 4.2914 |
3.4992 | 4.98 | 8500 | 4.2766 |
3.3394 | 5.28 | 9000 | 4.2968 |
3.324 | 5.57 | 9500 | 4.2965 |
3.3187 | 5.86 | 10000 | 4.2954 |
Framework versions
- Transformers 4.26.1
- Pytorch 1.11.0+cu113
- Datasets 2.13.0
- Tokenizers 0.13.3