<!-- 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-cbt-rarity-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: 4.3233
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 |
---|---|---|---|
6.6903 | 0.29 | 500 | 5.6411 |
5.3275 | 0.59 | 1000 | 5.2070 |
4.9793 | 0.88 | 1500 | 4.9578 |
4.7071 | 1.18 | 2000 | 4.8157 |
4.5592 | 1.47 | 2500 | 4.6921 |
4.4572 | 1.76 | 3000 | 4.5948 |
4.3319 | 2.06 | 3500 | 4.5170 |
4.1487 | 2.35 | 4000 | 4.4718 |
4.1144 | 2.65 | 4500 | 4.4073 |
4.0779 | 2.94 | 5000 | 4.3537 |
3.8613 | 3.23 | 5500 | 4.3448 |
3.8195 | 3.53 | 6000 | 4.3136 |
3.8016 | 3.82 | 6500 | 4.2823 |
3.6777 | 4.12 | 7000 | 4.2794 |
3.5343 | 4.41 | 7500 | 4.2711 |
3.5272 | 4.7 | 8000 | 4.2565 |
3.5136 | 5.0 | 8500 | 4.2434 |
3.3438 | 5.29 | 9000 | 4.2570 |
3.3322 | 5.58 | 9500 | 4.2559 |
3.3338 | 5.88 | 10000 | 4.2550 |
Framework versions
- Transformers 4.26.1
- Pytorch 1.11.0+cu113
- Datasets 2.13.0
- Tokenizers 0.13.3