G0515HMA15H
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: 0.1189
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 |
---|---|---|---|
3.2098 | 0.09 | 10 | 2.8516 |
2.5967 | 0.18 | 20 | 2.1281 |
1.6718 | 0.27 | 30 | 1.1013 |
0.6829 | 0.36 | 40 | 0.2924 |
0.2107 | 0.45 | 50 | 0.1645 |
0.1575 | 0.54 | 60 | 0.1545 |
0.1513 | 0.63 | 70 | 0.1509 |
0.1525 | 0.73 | 80 | 0.1491 |
0.145 | 0.82 | 90 | 0.1503 |
0.1484 | 0.91 | 100 | 0.1484 |
0.15 | 1.0 | 110 | 0.1488 |
0.1437 | 1.09 | 120 | 0.1483 |
0.145 | 1.18 | 130 | 0.1479 |
0.1459 | 1.27 | 140 | 0.1469 |
0.1487 | 1.36 | 150 | 0.1454 |
0.1418 | 1.45 | 160 | 0.1483 |
0.1435 | 1.54 | 170 | 0.1437 |
0.143 | 1.63 | 180 | 0.1416 |
0.1423 | 1.72 | 190 | 0.1391 |
0.1375 | 1.81 | 200 | 0.1352 |
0.138 | 1.9 | 210 | 0.1349 |
0.1345 | 1.99 | 220 | 0.1269 |
0.1297 | 2.08 | 230 | 0.1297 |
0.1267 | 2.18 | 240 | 0.1294 |
0.1264 | 2.27 | 250 | 0.1276 |
0.1255 | 2.36 | 260 | 0.1256 |
0.1249 | 2.45 | 270 | 0.1238 |
0.1191 | 2.54 | 280 | 0.1222 |
0.1169 | 2.63 | 290 | 0.1207 |
0.1163 | 2.72 | 300 | 0.1199 |
0.1199 | 2.81 | 310 | 0.1190 |
0.1226 | 2.9 | 320 | 0.1190 |
0.1203 | 2.99 | 330 | 0.1189 |
Framework versions
- Transformers 4.36.0.dev0
- Pytorch 2.1.2+cu121
- Datasets 2.14.6
- Tokenizers 0.14.0
Model tree for Litzy619/G0515HMA15H
Base model
google/gemma-2b
Finetuned
this model