base-multilingual-cased-sindhi-cpt
This model is a fine-tuned version of Kashif786/base-multilingual-cased-sindhi-cpt on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 2.6265
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: 5e-05
- train_batch_size: 128
- eval_batch_size: 128
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 256
- 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: cosine
- num_epochs: 1
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 7.6128 | 0.1423 | 3000 | 3.7017 |
| 6.4880 | 0.2846 | 6000 | 3.1180 |
| 5.9538 | 0.4269 | 9000 | 2.8777 |
| 5.6614 | 0.5691 | 12000 | 2.7585 |
| 5.5020 | 0.7114 | 15000 | 2.6584 |
| 5.4609 | 0.8537 | 18000 | 2.6361 |
| 5.4810 | 0.9960 | 21000 | 2.6258 |
| 5.4810 | 1.0 | 21085 | 2.6265 |
Framework versions
- Transformers 5.16.1
- Pytorch 2.8.0+cu129
- Datasets 5.0.1
- Tokenizers 0.23.2