metadata
license: mit
base_model: Narrativa/legal-longformer-base-4096-spanish
tags:
- generated_from_trainer
model-index:
- name: bert-leg-al-corpus
results: []
bert-leg-al-corpus
This model is a fine-tuned version of Narrativa/legal-longformer-base-4096-spanish on the None dataset. It achieves the following results on the evaluation set:
- Loss: 2.2219
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: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 20
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| No log | 1.0 | 120 | 2.7135 |
| No log | 2.0 | 240 | 2.5848 |
| No log | 3.0 | 360 | 2.5374 |
| No log | 4.0 | 480 | 2.4788 |
| 2.7765 | 5.0 | 600 | 2.4294 |
| 2.7765 | 6.0 | 720 | 2.3854 |
| 2.7765 | 7.0 | 840 | 2.3608 |
| 2.7765 | 8.0 | 960 | 2.3505 |
| 2.4729 | 9.0 | 1080 | 2.3189 |
| 2.4729 | 10.0 | 1200 | 2.3048 |
| 2.4729 | 11.0 | 1320 | 2.2949 |
| 2.4729 | 12.0 | 1440 | 2.2657 |
| 2.3559 | 13.0 | 1560 | 2.2393 |
| 2.3559 | 14.0 | 1680 | 2.2309 |
| 2.3559 | 15.0 | 1800 | 2.2160 |
| 2.3559 | 16.0 | 1920 | 2.2143 |
| 2.2843 | 17.0 | 2040 | 2.2282 |
| 2.2843 | 18.0 | 2160 | 2.2146 |
| 2.2843 | 19.0 | 2280 | 2.1929 |
| 2.2843 | 20.0 | 2400 | 2.1950 |
Framework versions
- Transformers 4.38.2
- Pytorch 2.2.1+cu121
- Datasets 2.18.0
- Tokenizers 0.15.2