V0414H3
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: 0.1402
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: 0.003
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 8
- total_train_batch_size: 64
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine_with_restarts
- lr_scheduler_warmup_steps: 80
- num_epochs: 3
- mixed_precision_training: Native AMP
Training results
Training Loss | Epoch | Step | Validation Loss |
---|---|---|---|
2.1861 | 0.05 | 10 | 1.1776 |
0.4648 | 0.09 | 20 | 0.1512 |
0.1485 | 0.14 | 30 | 0.1249 |
0.1167 | 0.18 | 40 | 0.1079 |
0.104 | 0.23 | 50 | 0.0901 |
0.1011 | 0.27 | 60 | 0.1029 |
0.095 | 0.32 | 70 | 0.0855 |
0.0966 | 0.36 | 80 | 0.0809 |
0.0818 | 0.41 | 90 | 0.0773 |
0.084 | 0.45 | 100 | 0.0750 |
0.0855 | 0.5 | 110 | 0.0741 |
0.0859 | 0.54 | 120 | 0.0722 |
0.0789 | 0.59 | 130 | 0.0810 |
0.0825 | 0.63 | 140 | 0.0757 |
0.0757 | 0.68 | 150 | 0.0720 |
0.0761 | 0.73 | 160 | 0.0825 |
0.0892 | 0.77 | 170 | 0.0815 |
0.0878 | 0.82 | 180 | 0.0781 |
0.0997 | 0.86 | 190 | 0.0707 |
0.0734 | 0.91 | 200 | 0.0773 |
0.096 | 0.95 | 210 | 0.0721 |
0.089 | 1.0 | 220 | 0.0768 |
0.0724 | 1.04 | 230 | 0.0762 |
0.0769 | 1.09 | 240 | 0.0754 |
0.0793 | 1.13 | 250 | 0.0739 |
0.0716 | 1.18 | 260 | 0.0777 |
0.0803 | 1.22 | 270 | 0.0756 |
0.0651 | 1.27 | 280 | 0.0723 |
0.0719 | 1.31 | 290 | 0.0672 |
0.0798 | 1.36 | 300 | 0.0821 |
0.0858 | 1.41 | 310 | 0.0871 |
0.0833 | 1.45 | 320 | 0.0736 |
0.0779 | 1.5 | 330 | 0.0741 |
0.0765 | 1.54 | 340 | 0.0713 |
0.0727 | 1.59 | 350 | 0.0659 |
0.0667 | 1.63 | 360 | 0.0836 |
0.097 | 1.68 | 370 | 0.0742 |
0.071 | 1.72 | 380 | 0.0663 |
0.0648 | 1.77 | 390 | 0.0662 |
0.091 | 1.81 | 400 | 0.0820 |
0.103 | 1.86 | 410 | 0.2671 |
2.8133 | 1.9 | 420 | 2.7663 |
2.1821 | 1.95 | 430 | 1.3153 |
1.0958 | 1.99 | 440 | 0.5246 |
0.4358 | 2.04 | 450 | 0.3359 |
0.3002 | 2.08 | 460 | 0.2346 |
0.2218 | 2.13 | 470 | 0.2145 |
0.2252 | 2.18 | 480 | 0.1891 |
0.1987 | 2.22 | 490 | 0.1758 |
0.1739 | 2.27 | 500 | 0.1732 |
0.1658 | 2.31 | 510 | 0.1604 |
0.1599 | 2.36 | 520 | 0.1548 |
0.1562 | 2.4 | 530 | 0.1527 |
0.1583 | 2.45 | 540 | 0.1514 |
0.1547 | 2.49 | 550 | 0.1484 |
0.1498 | 2.54 | 560 | 0.1516 |
0.1544 | 2.58 | 570 | 0.1477 |
0.1577 | 2.63 | 580 | 0.1435 |
0.1451 | 2.67 | 590 | 0.1428 |
0.1422 | 2.72 | 600 | 0.1415 |
0.1461 | 2.76 | 610 | 0.1412 |
0.1523 | 2.81 | 620 | 0.1409 |
0.1457 | 2.86 | 630 | 0.1402 |
0.1407 | 2.9 | 640 | 0.1401 |
0.1453 | 2.95 | 650 | 0.1402 |
0.1478 | 2.99 | 660 | 0.1402 |
Framework versions
- Transformers 4.36.0.dev0
- Pytorch 2.1.2+cu121
- Datasets 2.14.6
- Tokenizers 0.14.1
Model tree for Litzy619/V0414H3
Base model
microsoft/phi-2