gptneo-txt2ARXMLv1.6_3000
This model is a fine-tuned version of EleutherAI/gpt-neo-125m on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.1562
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: 0.0002
- train_batch_size: 1
- eval_batch_size: 1
- seed: 42
- distributed_type: multi-GPU
- gradient_accumulation_steps: 8
- total_train_batch_size: 8
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 1000
- num_epochs: 10
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 1.1777 | 0.99 | 59 | 1.1377 |
| 0.6024 | 1.99 | 119 | 0.6514 |
| 0.4513 | 2.99 | 179 | 0.5405 |
| 0.4204 | 3.99 | 239 | 0.4539 |
| 0.3151 | 4.99 | 299 | 0.4075 |
| 0.2056 | 6.0 | 359 | 0.3154 |
| 0.1129 | 7.0 | 419 | 0.2234 |
| 0.0871 | 8.0 | 479 | 0.1742 |
| 0.0764 | 8.99 | 538 | 0.1575 |
| 0.0548 | 9.85 | 590 | 0.1562 |
Framework versions
- Transformers 4.28.1
- Pytorch 2.0.0+cu118
- Datasets 2.12.0
- Tokenizers 0.13.3