metadata
license: apache-2.0
base_model: distilbert-base-uncased
tags:
- generated_from_trainer
model-index:
- name: finetuned-DSCS24-mitre-distilbert-base-uncased-fill-mask
results: []
finetuned-DSCS24-mitre-distilbert-base-uncased-fill-mask
This model is a fine-tuned version of distilbert-base-uncased on the None dataset. It achieves the following results on the evaluation set:
- Loss: 1.9549
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: 1e-05
- train_batch_size: 64
- eval_batch_size: 64
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- num_epochs: 20
- mixed_precision_training: Native AMP
Training results
Training Loss | Epoch | Step | Validation Loss |
---|---|---|---|
2.6456 | 1.0 | 65 | 2.3685 |
2.3833 | 2.0 | 130 | 2.2661 |
2.282 | 3.0 | 195 | 2.2728 |
2.1715 | 4.0 | 260 | 2.1675 |
2.1186 | 5.0 | 325 | 2.1354 |
2.0968 | 6.0 | 390 | 2.0318 |
2.0885 | 7.0 | 455 | 2.1233 |
2.0212 | 8.0 | 520 | 2.0152 |
1.9305 | 9.0 | 585 | 2.0134 |
1.9843 | 10.0 | 650 | 2.0334 |
1.9682 | 11.0 | 715 | 1.9611 |
1.9383 | 12.0 | 780 | 2.0051 |
1.9075 | 13.0 | 845 | 1.9790 |
1.9107 | 14.0 | 910 | 1.9532 |
1.9352 | 15.0 | 975 | 1.9677 |
1.9102 | 16.0 | 1040 | 1.9569 |
1.9065 | 17.0 | 1105 | 1.9143 |
1.8659 | 18.0 | 1170 | 1.9818 |
1.8807 | 19.0 | 1235 | 1.9321 |
1.9088 | 20.0 | 1300 | 1.9549 |
Framework versions
- Transformers 4.41.2
- Pytorch 2.3.0+cu121
- Datasets 2.19.2
- Tokenizers 0.19.1