metadata
base_model: meta-llama/Meta-Llama-3.1-8B-Instruct
datasets:
- GaetanMichelet/chat-60_ft_task-1
- GaetanMichelet/chat-120_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_120-samples_config-1
results: []
Llama-31-8B_task-1_120-samples_config-1
This model is a fine-tuned version of meta-llama/Meta-Llama-3.1-8B-Instruct on the GaetanMichelet/chat-60_ft_task-1 and the GaetanMichelet/chat-120_ft_task-1 datasets. It achieves the following results on the evaluation set:
- Loss: 1.2432
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: 8
- total_train_batch_size: 8
- 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 |
|---|---|---|---|
| 1.8587 | 1.0 | 11 | 1.8962 |
| 1.5582 | 2.0 | 22 | 1.5353 |
| 1.3116 | 3.0 | 33 | 1.3440 |
| 1.1103 | 4.0 | 44 | 1.2486 |
| 0.9015 | 5.0 | 55 | 1.2432 |
| 0.6339 | 6.0 | 66 | 1.3448 |
| 0.2953 | 7.0 | 77 | 1.6649 |
| 0.2611 | 8.0 | 88 | 1.9195 |
| 0.107 | 9.0 | 99 | 2.4669 |
| 0.0592 | 10.0 | 110 | 2.5539 |
| 0.06 | 11.0 | 121 | 2.4769 |
| 0.0548 | 12.0 | 132 | 2.9299 |
Framework versions
- PEFT 0.12.0
- Transformers 4.44.0
- Pytorch 2.1.2+cu121
- Datasets 2.20.0
- Tokenizers 0.19.1