gemini_chakma_distilbert
This model is a fine-tuned version of distilbert/distilbert-base-multilingual-cased on the None dataset. It achieves the following results on the evaluation set:
- Loss: 2.0421
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: 2e-05
- train_batch_size: 4
- eval_batch_size: 4
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 8
- 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: linear
- num_epochs: 20
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 2.7747 | 1.0 | 228 | 2.9578 |
| 2.4199 | 2.0 | 456 | 2.7325 |
| 2.241 | 3.0 | 684 | 2.6304 |
| 2.1286 | 4.0 | 912 | 2.4909 |
| 2.0423 | 5.0 | 1140 | 2.3890 |
| 1.9952 | 6.0 | 1368 | 2.4163 |
| 1.9284 | 7.0 | 1596 | 2.3252 |
| 1.8844 | 8.0 | 1824 | 2.2711 |
| 1.8333 | 9.0 | 2052 | 2.1887 |
| 1.8095 | 10.0 | 2280 | 2.2208 |
| 1.7648 | 11.0 | 2508 | 2.1549 |
| 1.7433 | 12.0 | 2736 | 2.1073 |
| 1.7116 | 13.0 | 2964 | 2.1316 |
| 1.7011 | 14.0 | 3192 | 2.0648 |
| 1.681 | 15.0 | 3420 | 2.0522 |
| 1.6659 | 16.0 | 3648 | 2.0799 |
| 1.6618 | 17.0 | 3876 | 2.0519 |
| 1.6376 | 18.0 | 4104 | 1.9956 |
| 1.6127 | 19.0 | 4332 | 2.0402 |
| 1.6214 | 20.0 | 4560 | 2.0421 |
Framework versions
- Transformers 4.56.1
- Pytorch 2.8.0+cu126
- Datasets 4.0.0
- Tokenizers 0.22.0
- Downloads last month
- 11