metadata
library_name: peft
license: llama3.2
base_model: meta-llama/Llama-3.2-3B
tags:
- generated_from_trainer
metrics:
- accuracy
model-index:
- name: LLama3-1B-finetuning-italian-version
results: []
LLama3-1B-finetuning-italian-version
This model is a fine-tuned version of meta-llama/Llama-3.2-3B on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.4550
- Model Preparation Time: 0.0066
- Accuracy: 0.8313
- F1 Macro: 0.8378
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: 16
- eval_batch_size: 16
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 32
- 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_ratio: 0.1
- num_epochs: 10
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Model Preparation Time | Accuracy | F1 Macro |
|---|---|---|---|---|---|---|
| 0.5588 | 1.0 | 368 | 0.5780 | 0.0066 | 0.7503 | 0.7548 |
| 0.4576 | 2.0 | 736 | 0.4660 | 0.0066 | 0.8048 | 0.8131 |
| 0.2993 | 3.0 | 1104 | 0.4513 | 0.0066 | 0.8177 | 0.8253 |
| 0.1717 | 4.0 | 1472 | 0.5759 | 0.0066 | 0.8020 | 0.8123 |
| 0.1154 | 5.0 | 1840 | 0.6625 | 0.0066 | 0.8136 | 0.8214 |
Framework versions
- PEFT 0.14.0
- Transformers 4.49.0
- Pytorch 2.5.1+cu124
- Datasets 3.3.1
- Tokenizers 0.21.0