G0515HMA22H
This model is a fine-tuned version of google/gemma-2b on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.1322
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: 80
- num_epochs: 3
- mixed_precision_training: Native AMP
Training results
Training Loss | Epoch | Step | Validation Loss |
---|---|---|---|
3.1841 | 0.09 | 10 | 2.8751 |
2.5107 | 0.18 | 20 | 1.9774 |
1.4768 | 0.27 | 30 | 0.8765 |
0.4892 | 0.36 | 40 | 0.2067 |
0.1718 | 0.45 | 50 | 0.1561 |
0.1522 | 0.54 | 60 | 0.1523 |
0.1497 | 0.63 | 70 | 0.1483 |
0.15 | 0.73 | 80 | 0.1496 |
0.1434 | 0.82 | 90 | 0.1502 |
0.1464 | 0.91 | 100 | 0.1485 |
0.1496 | 1.0 | 110 | 0.1545 |
0.1448 | 1.09 | 120 | 0.1500 |
0.1452 | 1.18 | 130 | 0.1482 |
0.1457 | 1.27 | 140 | 0.1470 |
0.1489 | 1.36 | 150 | 0.1466 |
0.1425 | 1.45 | 160 | 0.1505 |
0.1444 | 1.54 | 170 | 0.1477 |
0.1463 | 1.63 | 180 | 0.1462 |
0.1459 | 1.72 | 190 | 0.1485 |
0.1459 | 1.81 | 200 | 0.1476 |
0.1475 | 1.9 | 210 | 0.1470 |
0.1469 | 1.99 | 220 | 0.1461 |
0.1429 | 2.08 | 230 | 0.1453 |
0.1386 | 2.18 | 240 | 0.1431 |
0.14 | 2.27 | 250 | 0.1416 |
0.1406 | 2.36 | 260 | 0.1402 |
0.1379 | 2.45 | 270 | 0.1397 |
0.1349 | 2.54 | 280 | 0.1366 |
0.1358 | 2.63 | 290 | 0.1348 |
0.1319 | 2.72 | 300 | 0.1333 |
0.1337 | 2.81 | 310 | 0.1322 |
0.1322 | 2.9 | 320 | 0.1322 |
0.1335 | 2.99 | 330 | 0.1322 |
Framework versions
- Transformers 4.36.0.dev0
- Pytorch 2.1.2+cu121
- Datasets 2.14.6
- Tokenizers 0.14.0
Model tree for Litzy619/G0515HMA22H
Base model
google/gemma-2b
Finetuned
this model