bart-noised-with-gcd-babylon-dist
This model is a fine-tuned version of gayanin/bart-noised-with-gcd-dist on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.2281
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: 5e-05
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 10
- num_epochs: 3
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 0.2582 | 0.11 | 500 | 0.2439 |
| 0.2496 | 0.21 | 1000 | 0.2384 |
| 0.2469 | 0.32 | 1500 | 0.2428 |
| 0.2786 | 0.43 | 2000 | 0.2409 |
| 0.195 | 0.54 | 2500 | 0.2409 |
| 0.2615 | 0.64 | 3000 | 0.2352 |
| 0.2593 | 0.75 | 3500 | 0.2359 |
| 0.2472 | 0.86 | 4000 | 0.2340 |
| 0.2762 | 0.96 | 4500 | 0.2285 |
| 0.181 | 1.07 | 5000 | 0.2374 |
| 0.1963 | 1.18 | 5500 | 0.2345 |
| 0.1848 | 1.28 | 6000 | 0.2378 |
| 0.181 | 1.39 | 6500 | 0.2343 |
| 0.2063 | 1.5 | 7000 | 0.2299 |
| 0.1774 | 1.61 | 7500 | 0.2302 |
| 0.2058 | 1.71 | 8000 | 0.2267 |
| 0.2256 | 1.82 | 8500 | 0.2262 |
| 0.1661 | 1.93 | 9000 | 0.2265 |
| 0.1475 | 2.03 | 9500 | 0.2327 |
| 0.1229 | 2.14 | 10000 | 0.2338 |
| 0.1484 | 2.25 | 10500 | 0.2326 |
| 0.1679 | 2.35 | 11000 | 0.2326 |
| 0.1278 | 2.46 | 11500 | 0.2312 |
| 0.143 | 2.57 | 12000 | 0.2291 |
| 0.1331 | 2.68 | 12500 | 0.2284 |
| 0.1504 | 2.78 | 13000 | 0.2289 |
| 0.1308 | 2.89 | 13500 | 0.2293 |
| 0.1657 | 3.0 | 14000 | 0.2281 |
Framework versions
- Transformers 4.37.2
- Pytorch 2.1.2+cu121
- Datasets 2.16.1
- Tokenizers 0.15.1