O0430HMA17
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: 0.0188
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: 60
- num_epochs: 3
- mixed_precision_training: Native AMP
Training results
Training Loss | Epoch | Step | Validation Loss |
---|---|---|---|
2.2942 | 0.09 | 10 | 0.1808 |
0.1615 | 0.18 | 20 | 0.1583 |
0.1524 | 0.27 | 30 | 0.1564 |
0.1564 | 0.36 | 40 | 0.1529 |
0.1528 | 0.45 | 50 | 0.1525 |
0.1533 | 0.54 | 60 | 0.1504 |
0.1528 | 0.63 | 70 | 0.1483 |
0.147 | 0.73 | 80 | 0.1365 |
0.4162 | 0.82 | 90 | 0.2004 |
0.3136 | 0.91 | 100 | 0.0837 |
0.15 | 1.0 | 110 | 0.0849 |
0.0947 | 1.09 | 120 | 0.0721 |
0.1072 | 1.18 | 130 | 0.3448 |
0.0929 | 1.27 | 140 | 0.0710 |
0.7574 | 1.36 | 150 | 0.4213 |
0.1423 | 1.45 | 160 | 0.0615 |
0.0548 | 1.54 | 170 | 0.0528 |
0.0641 | 1.63 | 180 | 0.0572 |
0.0594 | 1.72 | 190 | 0.0471 |
0.0438 | 1.81 | 200 | 0.0419 |
0.0362 | 1.9 | 210 | 0.0342 |
0.0272 | 1.99 | 220 | 0.0235 |
0.0372 | 2.08 | 230 | 0.0306 |
0.0254 | 2.18 | 240 | 0.0238 |
0.0194 | 2.27 | 250 | 0.0227 |
0.0253 | 2.36 | 260 | 0.0218 |
0.0255 | 2.45 | 270 | 0.0208 |
0.0171 | 2.54 | 280 | 0.0208 |
0.0246 | 2.63 | 290 | 0.0204 |
0.0215 | 2.72 | 300 | 0.0197 |
0.019 | 2.81 | 310 | 0.0195 |
0.0205 | 2.9 | 320 | 0.0188 |
0.021 | 2.99 | 330 | 0.0188 |
Framework versions
- Transformers 4.36.0.dev0
- Pytorch 2.1.2+cu121
- Datasets 2.14.6
- Tokenizers 0.14.1
Model tree for Litzy619/O0430HMA17
Base model
allenai/OLMo-1B