mBART_billsum_desc_LT_nofreeze_model_en_NEU

This model is a fine-tuned version of facebook/mbart-large-50 on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 2.0750
• Rouge1: 0.4735
• Rouge2: 0.272
• Rougel: 0.3564
• Rougelsum: 0.3567
• Gen Len: 110.04

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: 2
• eval_batch_size: 2
• 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: 4
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum	Gen Len
5.8454	1.0	500	2.8617	0.0991	0.0444	0.0776	0.0783	124.38
2.4369	2.0	1000	2.2604	0.3651	0.2028	0.2747	0.2759	115.14
1.8154	3.0	1500	2.0983	0.4693	0.272	0.3581	0.358	113.78
1.5242	4.0	2000	2.0750	0.4735	0.272	0.3564	0.3567	110.04

Framework versions

• Transformers 4.52.4
• Pytorch 2.6.0+cu124
• Datasets 3.6.0
• Tokenizers 0.21.1