Llama-31-8B_task-1_120-samples_config-1_auto
This model is a fine-tuned version of meta-llama/Meta-Llama-3.1-8B-Instruct on the GaetanMichelet/chat-60_ft_task-1_auto and the GaetanMichelet/chat-120_ft_task-1_auto datasets.
It achieves the following results on the evaluation set:
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 |
1.2238 |
1.0 |
11 |
1.1612 |
0.8979 |
2.0 |
22 |
1.0152 |
0.8593 |
3.0 |
33 |
0.9378 |
0.7403 |
4.0 |
44 |
0.8861 |
0.5823 |
5.0 |
55 |
0.9116 |
0.4269 |
6.0 |
66 |
1.0325 |
0.2688 |
7.0 |
77 |
1.0847 |
0.1625 |
8.0 |
88 |
1.2475 |
0.0921 |
9.0 |
99 |
1.4958 |
0.0629 |
10.0 |
110 |
1.5466 |
0.0518 |
11.0 |
121 |
1.6494 |
Framework versions
- PEFT 0.12.0
- Transformers 4.44.0
- Pytorch 2.1.2+cu121
- Datasets 2.20.0
- Tokenizers 0.19.1