metadata
base_model: meta-llama/Meta-Llama-3-8B-Instruct
library_name: peft
license: llama3
tags:
- trl
- sft
- generated_from_trainer
model-index:
- name: results
results: []
results
This model is a fine-tuned version of meta-llama/Meta-Llama-3-8B-Instruct on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.4878
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: 1e-05
- train_batch_size: 3
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 12
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 20
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 0.6255 | 1.9665 | 88 | 0.6068 |
| 0.5655 | 3.9330 | 176 | 0.5801 |
| 0.5342 | 5.8994 | 264 | 0.5503 |
| 0.5044 | 7.8659 | 352 | 0.5312 |
| 0.48 | 9.8324 | 440 | 0.5123 |
| 0.4487 | 11.7989 | 528 | 0.5025 |
| 0.4325 | 13.7654 | 616 | 0.4943 |
| 0.4053 | 15.7318 | 704 | 0.4901 |
| 0.4218 | 17.6983 | 792 | 0.4892 |
| 0.4087 | 19.6648 | 880 | 0.4878 |
Framework versions
- PEFT 0.10.0
- Transformers 4.42.3
- Pytorch 2.3.1+cu121
- Datasets 2.18.0
- Tokenizers 0.19.1