metadata
library_name: peft
license: apache-2.0
base_model: unsloth/tinyllama-chat-bnb-4bit
tags:
- unsloth
- generated_from_trainer
model-index:
- name: englishTeluguModel
results: []
englishTeluguModel
This model is a fine-tuned version of unsloth/tinyllama-chat-bnb-4bit on the None dataset. It achieves the following results on the evaluation set:
- Loss: 4.8606
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 |
|---|---|---|---|
| 3.1943 | 0.4 | 2 | 5.0423 |
| 3.5 | 0.8 | 4 | 5.0423 |
| 4.6154 | 1.2 | 6 | 5.0423 |
| 3.1785 | 1.6 | 8 | 5.0423 |
| 4.2688 | 2.0 | 10 | 5.0344 |
| 3.2921 | 2.4 | 12 | 5.0084 |
| 3.3623 | 2.8 | 14 | 5.0084 |
| 3.35 | 3.2 | 16 | 4.9870 |
| 2.9667 | 3.6 | 18 | 4.9529 |
| 1.6306 | 4.0 | 20 | 4.9077 |
| 2.4324 | 4.4 | 22 | 4.8973 |
| 3.3228 | 4.8 | 24 | 4.8848 |
| 3.1761 | 5.2 | 26 | 4.8712 |
| 2.3414 | 5.6 | 28 | 4.8483 |
| 3.8609 | 6.0 | 30 | 4.8307 |
| 2.0469 | 6.4 | 32 | 4.8273 |
| 1.8496 | 6.8 | 34 | 4.8161 |
| 3.4417 | 7.2 | 36 | 4.8191 |
| 1.3278 | 7.6 | 38 | 4.8450 |
| 2.2355 | 8.0 | 40 | 4.8493 |
| 1.0422 | 8.4 | 42 | 4.8465 |
| 2.1946 | 8.8 | 44 | 4.8404 |
| 1.6153 | 9.2 | 46 | 4.8502 |
| 1.0659 | 9.6 | 48 | 4.8601 |
| 1.7748 | 10.0 | 50 | 4.8606 |
Framework versions
- PEFT 0.14.0
- Transformers 4.48.3
- Pytorch 2.6.0+cu124
- Datasets 3.3.1
- Tokenizers 0.21.0