telugu_colloquial_dataset
This model is a fine-tuned version of unsloth/tinyllama-chat-bnb-4bit on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 3.1747
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: 4
- eval_batch_size: 4
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 8
- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 2
- num_epochs: 10
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 6.344 | 0.8 | 2 | 7.5379 |
| 6.3794 | 1.4 | 4 | 7.5379 |
| 6.2252 | 2.0 | 6 | 7.5379 |
| 4.7958 | 2.8 | 8 | 7.3706 |
| 1.3834 | 3.4 | 10 | 4.3751 |
| 1.1045 | 4.0 | 12 | 5.1464 |
| 0.9427 | 4.8 | 14 | 4.9710 |
| 0.8191 | 5.4 | 16 | 4.5209 |
| 0.61 | 6.0 | 18 | 3.9120 |
| 0.5775 | 6.8 | 20 | 3.1747 |
Framework versions
- PEFT 0.14.0
- Transformers 4.48.3
- Pytorch 2.6.0+cu124
- Datasets 3.3.1
- Tokenizers 0.21.0
- Downloads last month
- 8
Inference Providers
NEW
This model isn't deployed by any Inference Provider.
🙋
Ask for provider support
Model tree for ruby-0/telugu_colloquial_dataset
Base model
unsloth/tinyllama-chat-bnb-4bit