G0428HMA6
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: 0.1059
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 |
---|---|---|---|
2.645 | 0.09 | 10 | 1.7359 |
1.1171 | 0.18 | 20 | 0.4457 |
0.2438 | 0.27 | 30 | 0.1612 |
0.1568 | 0.36 | 40 | 0.1498 |
0.1473 | 0.45 | 50 | 0.1478 |
0.1471 | 0.54 | 60 | 0.1482 |
0.1545 | 0.63 | 70 | 0.1474 |
0.1526 | 0.73 | 80 | 0.1488 |
0.1433 | 0.82 | 90 | 0.1479 |
0.1452 | 0.91 | 100 | 0.1482 |
0.1488 | 1.0 | 110 | 0.1496 |
0.1438 | 1.09 | 120 | 0.1489 |
0.145 | 1.18 | 130 | 0.1476 |
0.1453 | 1.27 | 140 | 0.1467 |
0.1482 | 1.36 | 150 | 0.1462 |
0.1408 | 1.45 | 160 | 0.1443 |
0.1411 | 1.54 | 170 | 0.1384 |
0.1312 | 1.63 | 180 | 0.1297 |
0.1321 | 1.72 | 190 | 0.1316 |
0.1246 | 1.81 | 200 | 0.1237 |
0.1232 | 1.9 | 210 | 0.1183 |
0.12 | 1.99 | 220 | 0.1173 |
0.1099 | 2.08 | 230 | 0.1167 |
0.1069 | 2.18 | 240 | 0.1131 |
0.1032 | 2.27 | 250 | 0.1125 |
0.1063 | 2.36 | 260 | 0.1125 |
0.1052 | 2.45 | 270 | 0.1108 |
0.1024 | 2.54 | 280 | 0.1087 |
0.0945 | 2.63 | 290 | 0.1081 |
0.0971 | 2.72 | 300 | 0.1076 |
0.103 | 2.81 | 310 | 0.1065 |
0.1022 | 2.9 | 320 | 0.1060 |
0.1039 | 2.99 | 330 | 0.1059 |
Framework versions
- Transformers 4.36.0.dev0
- Pytorch 2.1.2+cu121
- Datasets 2.14.6
- Tokenizers 0.14.1
Model tree for Litzy619/G0428HMA6
Base model
google/gemma-2b