metadata
base_model: meta-llama/Meta-Llama-3.1-8B
library_name: peft
license: llama3.1
tags:
- trl
- sft
- generated_from_trainer
model-index:
- name: fine-tune-llama-3.1
results: []
fine-tune-llama-3.1
This model is a fine-tuned version of meta-llama/Meta-Llama-3.1-8B on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 1.2970
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: 0.0002
- train_batch_size: 4
- eval_batch_size: 4
- seed: 42
- gradient_accumulation_steps: 6
- total_train_batch_size: 24
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.01
- training_steps: 100
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| No log | 0.0244 | 10 | 1.3896 |
| No log | 0.0487 | 20 | 1.3412 |
| 1.3817 | 0.0731 | 30 | 1.3219 |
| 1.3817 | 0.0975 | 40 | 1.3223 |
| 1.3009 | 0.1219 | 50 | 1.3352 |
| 1.3009 | 0.1462 | 60 | 1.3165 |
| 1.3009 | 0.1706 | 70 | 1.3069 |
| 1.2783 | 0.1950 | 80 | 1.3015 |
| 1.2783 | 0.2193 | 90 | 1.2987 |
| 1.2929 | 0.2437 | 100 | 1.2970 |
Framework versions
- PEFT 0.12.0
- Transformers 4.44.2
- Pytorch 2.0.0+nv23.05
- Datasets 2.15.0
- Tokenizers 0.19.1