finetune_t5_small_last
This model is a fine-tuned version of cointegrated/rut5-small on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 1.7367
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: 0.0004
- train_batch_size: 12
- eval_batch_size: 12
- seed: 42
- gradient_accumulation_steps: 12
- total_train_batch_size: 144
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 100
- num_epochs: 8
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 2.8842 | 0.54 | 150 | 2.1580 |
| 2.692 | 1.08 | 300 | 2.0038 |
| 2.6062 | 1.62 | 450 | 1.9259 |
| 2.5237 | 2.16 | 600 | 1.8800 |
| 2.4775 | 2.7 | 750 | 1.8504 |
| 2.4355 | 3.24 | 900 | 1.8220 |
| 2.4075 | 3.78 | 1050 | 1.8016 |
| 2.3813 | 4.33 | 1200 | 1.7818 |
| 2.3555 | 4.87 | 1350 | 1.7742 |
| 2.3521 | 5.41 | 1500 | 1.7602 |
| 2.3295 | 5.95 | 1650 | 1.7517 |
| 2.3185 | 6.49 | 1800 | 1.7444 |
| 2.2994 | 7.03 | 1950 | 1.7416 |
| 2.3105 | 7.57 | 2100 | 1.7367 |
Framework versions
- Transformers 4.36.2
- Pytorch 2.1.2+cu121
- Datasets 2.15.0
- Tokenizers 0.15.0