mbart-en-np-seqtoseq-sentence-translation

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

• Loss: 1.1896
• Bleu: 40.4595
• Gen Len: 10.288

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: 8
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 5

Training results

Training Loss	Epoch	Step	Validation Loss	Bleu	Gen Len
1.0147	1.0	1250	0.9876	40.1501	9.885
0.6038	2.0	2500	1.0122	40.728	10.113
0.3557	3.0	3750	1.0809	35.9297	10.844
0.2071	4.0	5000	1.1502	40.4318	10.28
0.1241	5.0	6250	1.1896	40.4595	10.288

Framework versions

• Transformers 4.42.4
• Pytorch 2.4.0+cu121
• Datasets 2.21.0
• Tokenizers 0.19.1