Llama3.1-8b-base-SFT-2024-09-19_LoRAs
This model is a fine-tuned version of meta-llama/Meta-Llama-3.1-8B on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 1.2570
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: 1e-05
- train_batch_size: 6
- eval_batch_size: 1
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: constant
- num_epochs: 1.5
Training results
Training Loss | Epoch | Step | Validation Loss |
---|---|---|---|
1.7838 | 0.0793 | 1000 | 1.5859 |
1.4315 | 0.1586 | 2000 | 1.4652 |
1.3679 | 0.2380 | 3000 | 1.4315 |
1.3328 | 0.3173 | 4000 | 1.4037 |
1.3026 | 0.3966 | 5000 | 1.3846 |
1.2827 | 0.4759 | 6000 | 1.3699 |
1.2656 | 0.5552 | 7000 | 1.3557 |
1.2541 | 0.6346 | 8000 | 1.3539 |
1.2443 | 0.7139 | 9000 | 1.3350 |
1.2324 | 0.7932 | 10000 | 1.3200 |
1.2136 | 0.8725 | 11000 | 1.3127 |
1.1994 | 0.9519 | 12000 | 1.2950 |
1.1891 | 1.0312 | 13000 | 1.2957 |
1.1631 | 1.1105 | 14000 | 1.2821 |
1.1466 | 1.1898 | 15000 | 1.2709 |
1.1472 | 1.2691 | 16000 | 1.2706 |
1.1401 | 1.3485 | 17000 | 1.2609 |
1.1446 | 1.4278 | 18000 | 1.2570 |
Framework versions
- PEFT 0.12.0
- Transformers 4.44.2
- Pytorch 2.0.1+cu118
- Datasets 3.0.0
- Tokenizers 0.19.1
- Downloads last month
- 6
Model tree for ccibeekeoc42/Llama3.1-8b-base-SFT-2024-09-19_LoRAs
Base model
meta-llama/Llama-3.1-8B