bart-base_fine_tuned_on_vnx_dev_by_hydra_1751811080_
This model is a fine-tuned version of facebook/bart-base on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 5.2110
- Rouge1: 41.1367
- Rouge2: 13.0681
- Rougel: 26.1238
- Rougelsum: 27.8081
- Gen Len: 20.6296
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: 16
- eval_batch_size: 16
- seed: 42
- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- num_epochs: 10
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum | Gen Len |
|---|---|---|---|---|---|---|---|---|
| No log | 1.0 | 166 | 5.6366 | 39.1563 | 12.3465 | 25.4499 | 26.956 | 19.7191 |
| No log | 2.0 | 332 | 5.4404 | 39.8879 | 12.1156 | 25.4926 | 27.059 | 20.5401 |
| No log | 3.0 | 498 | 5.3834 | 39.9383 | 12.3787 | 25.5092 | 27.077 | 20.2716 |
| 5.7108 | 4.0 | 664 | 5.3610 | 40.5839 | 12.8003 | 25.8965 | 27.5882 | 20.4938 |
| 5.7108 | 5.0 | 830 | 5.2411 | 40.6231 | 12.639 | 25.7687 | 27.4955 | 20.6698 |
| 5.7108 | 6.0 | 996 | 5.2553 | 41.0724 | 12.9644 | 26.0256 | 27.6656 | 20.3858 |
| 5.0784 | 7.0 | 1162 | 5.2598 | 41.1767 | 12.9214 | 26.1634 | 27.8755 | 20.642 |
| 5.0784 | 8.0 | 1328 | 5.2101 | 41.2257 | 13.0919 | 25.9916 | 27.7688 | 20.5895 |
| 5.0784 | 9.0 | 1494 | 5.2101 | 41.4749 | 13.1724 | 26.1424 | 27.9396 | 20.6111 |
| 4.88 | 10.0 | 1660 | 5.2110 | 41.1367 | 13.0681 | 26.1238 | 27.8081 | 20.6296 |
Framework versions
- Transformers 4.53.1
- Pytorch 2.7.1+cu128
- Datasets 3.6.0
- Tokenizers 0.21.2