<!-- 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-simple-wiki-mod-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: 4.3543
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 |
---|---|---|---|
6.6838 | 0.29 | 500 | 5.6277 |
5.3231 | 0.59 | 1000 | 5.1994 |
4.987 | 0.88 | 1500 | 4.9572 |
4.7151 | 1.17 | 2000 | 4.8128 |
4.5647 | 1.47 | 2500 | 4.7004 |
4.4618 | 1.76 | 3000 | 4.6135 |
4.3426 | 2.06 | 3500 | 4.5400 |
4.1605 | 2.35 | 4000 | 4.4888 |
4.1305 | 2.64 | 4500 | 4.4288 |
4.0903 | 2.94 | 5000 | 4.3762 |
3.8797 | 3.23 | 5500 | 4.3722 |
3.83 | 3.52 | 6000 | 4.3423 |
3.8158 | 3.82 | 6500 | 4.3083 |
3.6986 | 4.11 | 7000 | 4.3079 |
3.5427 | 4.4 | 7500 | 4.3022 |
3.5399 | 4.7 | 8000 | 4.2835 |
3.5248 | 4.99 | 8500 | 4.2710 |
3.352 | 5.28 | 9000 | 4.2862 |
3.3468 | 5.58 | 9500 | 4.2856 |
3.3441 | 5.87 | 10000 | 4.2850 |
Framework versions
- Transformers 4.26.1
- Pytorch 1.11.0+cu113
- Datasets 2.13.0
- Tokenizers 0.13.3