clinical_longformer_same_tokens_exp_gpu
This model is a fine-tuned version of allenai/longformer-base-4096 on the None dataset. It achieves the following results on the evaluation set:
- Loss: 1.1468
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: 2e-05
- train_batch_size: 4
- eval_batch_size: 32
- seed: 42
- gradient_accumulation_steps: 64
- total_train_batch_size: 256
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 1500
- num_epochs: 3
Training results
Training Loss | Epoch | Step | Validation Loss |
---|---|---|---|
3.0528 | 0.07 | 65 | 2.6194 |
3.2272 | 0.14 | 130 | 2.4923 |
2.8624 | 0.21 | 195 | 2.3739 |
2.552 | 0.29 | 260 | 2.2678 |
2.7232 | 0.36 | 325 | 2.1829 |
2.361 | 0.43 | 390 | 2.0939 |
2.6223 | 0.5 | 455 | 2.0165 |
2.2567 | 0.57 | 520 | 1.9490 |
2.1168 | 0.64 | 585 | 1.8869 |
2.1214 | 0.71 | 650 | 1.8269 |
2.3123 | 0.79 | 715 | 1.7750 |
2.2535 | 0.86 | 780 | 1.7338 |
1.8926 | 0.93 | 845 | 1.6785 |
1.7711 | 1.0 | 910 | 1.6277 |
1.7578 | 1.07 | 975 | 1.5990 |
2.0243 | 1.14 | 1040 | 1.5484 |
2.2152 | 1.21 | 1105 | 1.5203 |
1.6819 | 1.28 | 1170 | 1.4809 |
1.5537 | 1.36 | 1235 | 1.4518 |
1.574 | 1.43 | 1300 | 1.4189 |
1.6772 | 1.5 | 1365 | 1.3917 |
1.6269 | 1.57 | 1430 | 1.3563 |
1.5375 | 1.64 | 1495 | 1.3363 |
1.5477 | 1.71 | 1560 | 1.3118 |
1.4559 | 1.78 | 1625 | 1.2857 |
1.4579 | 1.86 | 1690 | 1.2716 |
1.4705 | 1.93 | 1755 | 1.2587 |
1.4205 | 2.0 | 1820 | 1.2346 |
1.423 | 2.07 | 1885 | 1.2277 |
1.3852 | 2.14 | 1950 | 1.2123 |
1.3362 | 2.21 | 2015 | 1.2078 |
1.2793 | 2.28 | 2080 | 1.1854 |
1.3126 | 2.36 | 2145 | 1.1806 |
1.28 | 2.43 | 2210 | 1.1731 |
1.3789 | 2.5 | 2275 | 1.1646 |
1.3572 | 2.57 | 2340 | 1.1648 |
1.3398 | 2.64 | 2405 | 1.1539 |
1.2831 | 2.71 | 2470 | 1.1518 |
1.3269 | 2.78 | 2535 | 1.1488 |
1.2208 | 2.86 | 2600 | 1.1478 |
1.2955 | 2.93 | 2665 | 1.1421 |
1.2286 | 3.0 | 2730 | 1.1468 |
Framework versions
- Transformers 4.35.2
- Pytorch 2.1.1+cu121
- Datasets 2.15.0
- Tokenizers 0.15.0
- Downloads last month
- 0