tapt_helpfulness_base_pretraining_model
This model is a fine-tuned version of roberta-base on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 1.4502
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: 32
- eval_batch_size: 32
- seed: 42
- gradient_accumulation_steps: 11
- total_train_batch_size: 352
- optimizer: Adam with betas=(0.9,0.98) and epsilon=1e-06
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.06
- num_epochs: 10
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 1.9099 | 0.97 | 27 | 1.6497 |
| 1.716 | 1.98 | 55 | 1.6088 |
| 1.6549 | 2.99 | 83 | 1.5624 |
| 1.6585 | 3.97 | 110 | 1.5455 |
| 1.557 | 4.98 | 138 | 1.5446 |
| 1.5142 | 5.99 | 166 | 1.5057 |
| 1.4788 | 7.0 | 194 | 1.4934 |
| 1.5057 | 7.97 | 221 | 1.4714 |
| 1.4232 | 8.98 | 249 | 1.4541 |
| 1.3778 | 9.74 | 270 | 1.4498 |
Framework versions
- Transformers 4.38.2
- Pytorch 2.2.1+cu121
- Datasets 2.18.0
- Tokenizers 0.15.2