metadata
license: llama3.1
base_model: meta-llama/Llama-3.1-8B-Instruct
tags:
- trl
- sft
- generated_from_trainer
library_name: peft
model-index:
- name: Llama-3.1-8B-Instruct_FT
results: []
Llama-3.1-8B-Instruct_FT
This model is a fine-tuned version of meta-llama/Llama-3.1-8B-Instruct on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.8396
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: 0.0001
- train_batch_size: 2
- eval_batch_size: 2
- seed: 42
- gradient_accumulation_steps: 8
- total_train_batch_size: 16
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: constant_with_warmup
- lr_scheduler_warmup_steps: 50
- training_steps: 100
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 4.7739 | 0.0170 | 10 | 4.0561 |
| 2.8826 | 0.0340 | 20 | 2.7339 |
| 2.1963 | 0.0510 | 30 | 1.9088 |
| 0.7378 | 0.0679 | 40 | 0.9899 |
| 1.1619 | 0.0849 | 50 | 0.8700 |
| 1.2501 | 0.1019 | 60 | 0.8633 |
| 1.3083 | 0.1189 | 70 | 0.8459 |
| 0.7243 | 0.1359 | 80 | 0.8478 |
| 1.2093 | 0.1529 | 90 | 0.8476 |
| 0.5714 | 0.1699 | 100 | 0.8396 |
Framework versions
- PEFT 0.10.0
- Transformers 4.44.0
- Pytorch 2.5.1+cu121
- Datasets 3.2.0
- Tokenizers 0.19.1