V0417MADP10
This model is a fine-tuned version of microsoft/phi-2 on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.0710
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: 0.0003
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 16
- total_train_batch_size: 128
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine_with_restarts
- lr_scheduler_warmup_steps: 60
- num_epochs: 3
- mixed_precision_training: Native AMP
Training results
Training Loss | Epoch | Step | Validation Loss |
---|---|---|---|
5.3469 | 0.09 | 10 | 2.3130 |
3.8736 | 0.18 | 20 | 0.9229 |
1.8292 | 0.27 | 30 | 0.1605 |
0.8383 | 0.36 | 40 | 0.1460 |
0.2009 | 0.45 | 50 | 0.1369 |
0.1511 | 0.54 | 60 | 0.1310 |
0.1462 | 0.63 | 70 | 0.1210 |
0.1384 | 0.73 | 80 | 0.1111 |
0.1187 | 0.82 | 90 | 0.1039 |
0.1094 | 0.91 | 100 | 0.0909 |
0.1087 | 1.0 | 110 | 0.0858 |
0.102 | 1.09 | 120 | 0.0930 |
0.1075 | 1.18 | 130 | 0.0894 |
0.1001 | 1.27 | 140 | 0.0791 |
0.1156 | 1.36 | 150 | 0.0952 |
0.1109 | 1.45 | 160 | 0.0944 |
0.1065 | 1.54 | 170 | 0.0892 |
0.0973 | 1.63 | 180 | 0.0803 |
0.0887 | 1.72 | 190 | 0.0758 |
0.0943 | 1.81 | 200 | 0.0742 |
0.0894 | 1.9 | 210 | 0.0834 |
0.0905 | 1.99 | 220 | 0.0797 |
0.0822 | 2.08 | 230 | 0.0764 |
0.0855 | 2.18 | 240 | 0.0741 |
0.0785 | 2.27 | 250 | 0.0740 |
0.0803 | 2.36 | 260 | 0.0719 |
0.0809 | 2.45 | 270 | 0.0737 |
0.0771 | 2.54 | 280 | 0.0730 |
0.073 | 2.63 | 290 | 0.0715 |
0.0768 | 2.72 | 300 | 0.0711 |
0.0782 | 2.81 | 310 | 0.0712 |
0.0732 | 2.9 | 320 | 0.0709 |
0.0816 | 2.99 | 330 | 0.0710 |
Framework versions
- Transformers 4.36.0.dev0
- Pytorch 2.2.2+cu121
- Datasets 2.18.0
- Tokenizers 0.14.1
Model tree for Litzy619/V0417MADP10
Base model
microsoft/phi-2