metadata
library_name: transformers
license: llama3.1
base_model: meta-llama/Llama-3.1-8B-Instruct
tags:
- alignment-handbook
- trl
- sft
- generated_from_trainer
- trl
- sft
- generated_from_trainer
model-index:
- name: llama_3_o1_5_full
results: []
llama_3_o1_5_full
This model is a fine-tuned version of meta-llama/Llama-3.1-8B-Instruct on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.6189
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: 2e-05
- train_batch_size: 4
- eval_batch_size: 4
- seed: 42
- distributed_type: multi-GPU
- num_devices: 4
- gradient_accumulation_steps: 2
- total_train_batch_size: 32
- total_eval_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: 1
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 0.7411 | 0.0824 | 100 | 0.7033 |
| 0.7256 | 0.1648 | 200 | 0.7274 |
| 0.7119 | 0.2472 | 300 | 0.7168 |
| 0.7061 | 0.3296 | 400 | 0.7045 |
| 0.6921 | 0.4120 | 500 | 0.6886 |
| 0.6667 | 0.4944 | 600 | 0.6813 |
| 0.6572 | 0.5768 | 700 | 0.6667 |
| 0.6404 | 0.6593 | 800 | 0.6457 |
| 0.625 | 0.7417 | 900 | 0.6345 |
| 0.6092 | 0.8241 | 1000 | 0.6246 |
| 0.6073 | 0.9065 | 1100 | 0.6203 |
| 0.5998 | 0.9889 | 1200 | 0.6190 |
Framework versions
- Transformers 4.44.2
- Pytorch 2.4.1+cu121
- Datasets 3.0.0
- Tokenizers 0.19.1