<!-- 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-guten-rarity-all-5k-2p5k
This model is a fine-tuned version of gpt2 on the generator dataset. It achieves the following results on the evaluation set:
- Loss: 3.1866
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 |
---|---|---|---|
6.7208 | 0.3 | 500 | 5.6450 |
5.3608 | 0.59 | 1000 | 5.1983 |
5.0083 | 0.89 | 1500 | 4.9525 |
4.7373 | 1.19 | 2000 | 4.8032 |
4.5839 | 1.48 | 2500 | 4.6775 |
4.47 | 1.78 | 3000 | 4.5693 |
4.3221 | 2.08 | 3500 | 4.4945 |
4.1498 | 2.37 | 4000 | 4.4429 |
4.1253 | 2.67 | 4500 | 4.3840 |
4.0837 | 2.97 | 5000 | 4.3308 |
3.8433 | 3.26 | 5500 | 4.3330 |
3.814 | 3.56 | 6000 | 4.2933 |
3.8137 | 3.86 | 6500 | 4.2625 |
3.6539 | 4.15 | 7000 | 4.2653 |
3.5382 | 4.45 | 7500 | 4.2571 |
3.5335 | 4.75 | 8000 | 4.2415 |
3.4835 | 5.04 | 8500 | 4.2410 |
3.3457 | 5.34 | 9000 | 4.2458 |
3.3322 | 5.64 | 9500 | 4.2436 |
3.3408 | 5.93 | 10000 | 4.2430 |
Framework versions
- Transformers 4.26.1
- Pytorch 1.11.0+cu113
- Datasets 2.13.0
- Tokenizers 0.13.3