gpt2_medium_AR_unigram_32768_parallel3_42
This model was trained from scratch on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 3.1090
- Accuracy: 0.4240
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: 5e-05
- train_batch_size: 64
- eval_batch_size: 64
- seed: 42
- distributed_type: multi-GPU
- num_devices: 2
- total_train_batch_size: 128
- total_eval_batch_size: 128
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- lr_scheduler_warmup_steps: 1000
- num_epochs: 10
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|
| 3.5332 | 1.0 | 29228 | 3.4832 | 0.3760 |
| 3.3706 | 2.0 | 58456 | 3.3357 | 0.3924 |
| 3.2785 | 3.0 | 87684 | 3.2656 | 0.4011 |
| 3.2184 | 4.0 | 116912 | 3.2214 | 0.4063 |
| 3.1664 | 5.0 | 146140 | 3.1894 | 0.4111 |
| 3.1251 | 6.0 | 175368 | 3.1643 | 0.4144 |
| 3.0843 | 7.0 | 204596 | 3.1442 | 0.4175 |
| 3.045 | 8.0 | 233824 | 3.1277 | 0.4202 |
| 3.0083 | 9.0 | 263052 | 3.1150 | 0.4222 |
| 2.9734 | 10.0 | 292280 | 3.1090 | 0.4240 |
Framework versions
- Transformers 4.41.2
- Pytorch 2.8.0+cu128
- Datasets 4.1.1
- Tokenizers 0.19.1
