O0428HMA15
This model is a fine-tuned version of allenai/OLMo-1B on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.1427
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: 100
- num_epochs: 3
- mixed_precision_training: Native AMP
Training results
Training Loss | Epoch | Step | Validation Loss |
---|---|---|---|
1.5625 | 0.09 | 10 | 0.2337 |
0.1706 | 0.18 | 20 | 0.1519 |
0.1507 | 0.27 | 30 | 0.1728 |
0.1576 | 0.36 | 40 | 0.1556 |
0.1494 | 0.45 | 50 | 0.1485 |
0.1503 | 0.54 | 60 | 0.1484 |
0.1504 | 0.63 | 70 | 0.1474 |
0.1504 | 0.73 | 80 | 0.1569 |
0.1468 | 0.82 | 90 | 0.1503 |
0.1497 | 0.91 | 100 | 0.1497 |
0.153 | 1.0 | 110 | 0.1512 |
0.1464 | 1.09 | 120 | 0.1479 |
0.1468 | 1.18 | 130 | 0.1521 |
0.1486 | 1.27 | 140 | 0.1491 |
0.1517 | 1.36 | 150 | 0.1495 |
0.1469 | 1.45 | 160 | 0.1515 |
0.1479 | 1.54 | 170 | 0.1503 |
0.1491 | 1.63 | 180 | 0.1471 |
0.148 | 1.72 | 190 | 0.1480 |
0.1456 | 1.81 | 200 | 0.1488 |
0.1504 | 1.9 | 210 | 0.1493 |
0.1468 | 1.99 | 220 | 0.1490 |
0.1471 | 2.08 | 230 | 0.1460 |
0.1413 | 2.18 | 240 | 0.1468 |
0.143 | 2.27 | 250 | 0.1468 |
0.1435 | 2.36 | 260 | 0.1476 |
0.1426 | 2.45 | 270 | 0.1461 |
0.1398 | 2.54 | 280 | 0.1447 |
0.1402 | 2.63 | 290 | 0.1446 |
0.1426 | 2.72 | 300 | 0.1424 |
0.1421 | 2.81 | 310 | 0.1428 |
0.1415 | 2.9 | 320 | 0.1428 |
0.1419 | 2.99 | 330 | 0.1427 |
Framework versions
- Transformers 4.36.0.dev0
- Pytorch 2.1.2+cu121
- Datasets 2.14.6
- Tokenizers 0.14.1
Model tree for Litzy619/O0428HMA15
Base model
allenai/OLMo-1B