IndoT5-LoRA-R8-LR0.0003-Ep10
This model is a fine-tuned version of Wikidepia/IndoT5-base on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.5073
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: 0.0003
- train_batch_size: 64
- eval_batch_size: 64
- seed: 42
- optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: constant
- num_epochs: 10
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 0.9111 | 1.0 | 188 | 0.8518 |
| 0.7769 | 2.0 | 376 | 0.7935 |
| 0.6305 | 3.0 | 564 | 0.5751 |
| 0.5439 | 4.0 | 752 | 0.5503 |
| 0.6083 | 5.0 | 940 | 0.5318 |
| 0.5484 | 6.0 | 1128 | 0.5256 |
| 0.5819 | 7.0 | 1316 | 0.5279 |
| 0.5411 | 8.0 | 1504 | 0.5208 |
| 0.5178 | 9.0 | 1692 | 0.5225 |
| 0.5 | 10.0 | 1880 | 0.5073 |
Framework versions
- PEFT 0.18.1
- Transformers 4.57.6
- Pytorch 2.10.0+cu128
- Datasets 4.5.0
- Tokenizers 0.22.2
- Downloads last month
- -
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support
Model tree for raniaa17/IndoT5-LoRA-R8-LR0.0003-Ep10
Base model
Wikidepia/IndoT5-base