llama3-8B_MT
This model is a fine-tuned version of meta-llama/Meta-Llama-3-8B on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.6988
- Accuracy: 0.7817
- Precision: 0.7394
- Recall: 0.87
- F1 score: 0.7994
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: 16
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 5
Training results
Training Loss | Epoch | Step | Accuracy | F1 score | Precision | Recall | Validation Loss |
---|---|---|---|---|---|---|---|
0.7787 | 0.5 | 200 | 0.7533 | 0.7605 | 0.7390 | 0.7833 | 0.6041 |
0.5489 | 1.0 | 400 | 0.73 | 0.7484 | 0.7006 | 0.8033 | 0.5793 |
0.3894 | 1.5 | 600 | 0.6433 | 0.7214 | 0.5919 | 0.9233 | 0.7408 |
0.3638 | 2.0 | 800 | 0.7533 | 0.7817 | 0.7011 | 0.8833 | 0.5249 |
0.2665 | 2.5 | 1000 | 0.4719 | 0.8 | 0.7744 | 0.8467 | 0.8089 |
0.2587 | 3.0 | 1200 | 0.4709 | 0.7917 | 0.7660 | 0.84 | 0.8013 |
0.1716 | 3.5 | 1400 | 0.4906 | 0.8083 | 0.7955 | 0.83 | 0.8124 |
0.1672 | 4.0 | 1600 | 0.8441 | 0.7083 | 0.6478 | 0.9133 | 0.7580 |
0.0866 | 4.5 | 1800 | 0.6659 | 0.7917 | 0.7522 | 0.87 | 0.8068 |
0.0822 | 5.0 | 2000 | 0.6988 | 0.7817 | 0.7394 | 0.87 | 0.7994 |
Framework versions
- PEFT 0.11.1
- Transformers 4.44.2
- Pytorch 2.3.0+cu121
- Datasets 2.19.1
- Tokenizers 0.19.1
- Downloads last month
- 10
Model tree for rishavranaut/llama3-8B_MT
Base model
meta-llama/Meta-Llama-3-8B