my_awesome_opus_books_model_mbart

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: 1.9794
• Bleu: 21.882
• Gen Len: 33.9978

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 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	Bleu	Gen Len
1.902	1.0	4674	1.8273	21.0918	34.6839
1.6234	2.0	9348	1.7442	22.2418	34.2832
1.4264	3.0	14022	1.7224	22.4991	34.2291
1.2445	4.0	18696	1.7395	22.5686	34.1969
1.1011	5.0	23370	1.7810	22.423	33.968
0.9834	6.0	28044	1.8191	22.356	33.9389
0.8978	7.0	32718	1.8680	22.1928	34.1662
0.7984	8.0	37392	1.9160	22.1593	34.27
0.7453	9.0	42066	1.9513	21.8321	33.8822
0.7035	10.0	46740	1.9794	21.882	33.9978

Framework versions

• Transformers 4.52.4
• Pytorch 2.7.1+cu126
• Datasets 3.6.0
• Tokenizers 0.21.1