metadata
base_model: meta-llama/Meta-Llama-3.1-8B-Instruct
datasets:
- GaetanMichelet/chat-60_ft_task-1
library_name: peft
license: llama3.1
tags:
- alignment-handbook
- trl
- sft
- generated_from_trainer
model-index:
- name: Llama-31-8B_task-1_60-samples_config-2
results: []
Llama-31-8B_task-1_60-samples_config-2
This model is a fine-tuned version of meta-llama/Meta-Llama-3.1-8B-Instruct on the GaetanMichelet/chat-60_ft_task-1 dataset. It achieves the following results on the evaluation set:
- Loss: 1.2880
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: 1
- eval_batch_size: 1
- seed: 42
- distributed_type: multi-GPU
- gradient_accumulation_steps: 16
- total_train_batch_size: 16
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 50
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 2.1668 | 0.6957 | 2 | 2.0610 |
| 2.1104 | 1.7391 | 5 | 1.8653 |
| 1.6856 | 2.7826 | 8 | 1.5622 |
| 1.4548 | 3.8261 | 11 | 1.4691 |
| 1.3644 | 4.8696 | 14 | 1.3791 |
| 1.2755 | 5.9130 | 17 | 1.3224 |
| 1.1243 | 6.9565 | 20 | 1.2880 |
| 1.0234 | 8.0 | 23 | 1.3072 |
| 0.933 | 8.6957 | 25 | 1.3176 |
| 0.7437 | 9.7391 | 28 | 1.4185 |
| 0.6589 | 10.7826 | 31 | 1.4576 |
| 0.4603 | 11.8261 | 34 | 1.5915 |
| 0.3502 | 12.8696 | 37 | 1.7686 |
| 0.1758 | 13.9130 | 40 | 1.9850 |
Framework versions
- PEFT 0.12.0
- Transformers 4.44.0
- Pytorch 2.1.2+cu121
- Datasets 2.20.0
- Tokenizers 0.19.1