metadata
library_name: peft
license: llama3.2
base_model: meta-llama/Llama-3.2-1B
tags:
- generated_from_trainer
model-index:
- name: Llama1B_LoRA_stsb
results: []
Llama1B_LoRA_stsb
This model is a fine-tuned version of meta-llama/Llama-3.2-1B on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.4743
- Pearson: 0.8890
- Spearman: 0.8899
- Combined Score: 0.8895
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: 3e-05
- train_batch_size: 16
- eval_batch_size: 8
- 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
- num_epochs: 10
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Pearson | Spearman | Combined Score |
|---|---|---|---|---|---|---|
| 8.6438 | 1.0 | 180 | 0.9095 | 0.7858 | 0.7921 | 0.7889 |
| 1.1093 | 2.0 | 360 | 0.6337 | 0.8518 | 0.8544 | 0.8531 |
| 0.8425 | 3.0 | 540 | 0.5438 | 0.8710 | 0.8729 | 0.8719 |
| 0.7033 | 4.0 | 720 | 0.5221 | 0.8772 | 0.8787 | 0.8779 |
| 0.6368 | 5.0 | 900 | 0.5185 | 0.8802 | 0.8827 | 0.8814 |
| 0.5775 | 6.0 | 1080 | 0.4911 | 0.8871 | 0.8884 | 0.8878 |
| 0.5258 | 7.0 | 1260 | 0.4864 | 0.8873 | 0.8884 | 0.8878 |
| 0.4999 | 8.0 | 1440 | 0.4874 | 0.8891 | 0.8901 | 0.8896 |
| 0.4644 | 9.0 | 1620 | 0.4783 | 0.8883 | 0.8892 | 0.8888 |
| 0.4409 | 10.0 | 1800 | 0.4743 | 0.8890 | 0.8899 | 0.8895 |
Framework versions
- PEFT 0.14.0
- Transformers 4.47.1
- Pytorch 2.5.1+cu121
- Datasets 3.2.0
- Tokenizers 0.21.0