metadata
license: apache-2.0
base_model: facebook/bart-large
tags:
- generated_from_trainer
datasets:
- arrow
model-index:
- name: LexLM_Longformer_BART_hybrid_V1
results: []
LexLM_Longformer_BART_hybrid_V1
This model is a fine-tuned version of facebook/bart-large on the arrow dataset. It achieves the following results on the evaluation set:
- Loss: 2.0869
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_ratio: 0.1
- num_epochs: 40
Training results
Training Loss | Epoch | Step | Validation Loss |
---|---|---|---|
4.0806 | 1.0 | 69 | 2.3843 |
2.4153 | 2.0 | 138 | 2.1901 |
2.1675 | 3.0 | 207 | 2.1073 |
2.024 | 4.0 | 276 | 2.0852 |
1.8995 | 5.0 | 345 | 2.0690 |
1.7917 | 6.0 | 414 | 2.0499 |
1.7063 | 7.0 | 483 | 2.0448 |
1.6201 | 8.0 | 552 | 2.0468 |
1.5533 | 9.0 | 621 | 2.0546 |
1.4816 | 10.0 | 690 | 2.0826 |
1.4184 | 11.0 | 759 | 2.0995 |
1.3638 | 12.0 | 828 | 2.0869 |
Framework versions
- Transformers 4.40.1
- Pytorch 2.2.1+cu121
- Datasets 2.17.1
- Tokenizers 0.19.1