<!-- 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-mod-datatsets-rarity-all-iorder-no-cut-repetition
This model is a fine-tuned version of gpt2 on the generator dataset. It achieves the following results on the evaluation set:
- Loss: 3.2189
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 |
---|---|---|---|
6.734 | 0.3 | 500 | 5.7011 |
5.3923 | 0.6 | 1000 | 5.2709 |
5.0404 | 0.9 | 1500 | 5.0055 |
4.7561 | 1.21 | 2000 | 4.8552 |
4.6189 | 1.51 | 2500 | 4.7257 |
4.5064 | 1.81 | 3000 | 4.6217 |
4.3497 | 2.11 | 3500 | 4.5593 |
4.1924 | 2.41 | 4000 | 4.5086 |
4.1669 | 2.71 | 4500 | 4.4446 |
4.102 | 3.01 | 5000 | 4.4099 |
3.8642 | 3.32 | 5500 | 4.4021 |
3.8619 | 3.62 | 6000 | 4.3641 |
3.8392 | 3.92 | 6500 | 4.3356 |
3.6347 | 4.22 | 7000 | 4.3605 |
3.5759 | 4.52 | 7500 | 4.3424 |
3.5613 | 4.82 | 8000 | 4.3281 |
3.4782 | 5.12 | 8500 | 4.3392 |
3.3739 | 5.42 | 9000 | 4.3409 |
3.3737 | 5.73 | 9500 | 4.3415 |
Framework versions
- Transformers 4.26.1
- Pytorch 1.11.0+cu113
- Datasets 2.13.0
- Tokenizers 0.13.3