V0417MADP2
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: 0.1574
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 |
---|---|---|---|
8.5044 | 0.09 | 10 | 3.0469 |
7.1174 | 0.18 | 20 | 2.9171 |
6.0751 | 0.27 | 30 | 2.6390 |
4.2708 | 0.36 | 40 | 2.0163 |
2.4666 | 0.45 | 50 | 1.5118 |
1.3427 | 0.54 | 60 | 0.8933 |
0.5622 | 0.63 | 70 | 0.4358 |
0.2583 | 0.73 | 80 | 0.2698 |
0.2135 | 0.82 | 90 | 0.2154 |
0.1981 | 0.91 | 100 | 0.1957 |
0.1955 | 1.0 | 110 | 0.1945 |
0.2021 | 1.09 | 120 | 0.2029 |
0.1932 | 1.18 | 130 | 0.1893 |
0.1726 | 1.27 | 140 | 0.1965 |
0.1813 | 1.36 | 150 | 0.1825 |
0.1865 | 1.45 | 160 | 0.1699 |
0.1787 | 1.54 | 170 | 0.1609 |
0.1634 | 1.63 | 180 | 0.1666 |
0.1673 | 1.72 | 190 | 0.1703 |
0.2204 | 1.81 | 200 | 0.1684 |
0.1751 | 1.9 | 210 | 0.1619 |
0.1656 | 1.99 | 220 | 0.1665 |
0.1717 | 2.08 | 230 | 0.1583 |
0.1664 | 2.18 | 240 | 0.1635 |
0.1682 | 2.27 | 250 | 0.1628 |
0.1729 | 2.36 | 260 | 0.1635 |
0.1703 | 2.45 | 270 | 0.1622 |
0.168 | 2.54 | 280 | 0.1578 |
0.1588 | 2.63 | 290 | 0.1564 |
0.1554 | 2.72 | 300 | 0.1571 |
0.1566 | 2.81 | 310 | 0.1573 |
0.1602 | 2.9 | 320 | 0.1572 |
0.1587 | 2.99 | 330 | 0.1574 |
Framework versions
- Transformers 4.36.0.dev0
- Pytorch 2.1.2+cu121
- Datasets 2.14.6
- Tokenizers 0.14.1
Model tree for Litzy619/V0417MADP2
Base model
microsoft/phi-2