BARTBana_Translation_v3

This model is a fine-tuned version of IAmSkyDra/BARTBana_v3 on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.4571
• Sacrebleu: 11.0566

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: 64
• eval_batch_size: 64
• 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: 10
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Sacrebleu
0.7013	1.0	742	0.6063	6.2147
0.6067	2.0	1484	0.5359	8.3509
0.5273	3.0	2226	0.5025	9.5105
0.5043	4.0	2968	0.4846	10.0064
0.4699	5.0	3710	0.4747	10.4437
0.4558	6.0	4452	0.4658	10.6266
0.4296	7.0	5194	0.4613	10.8692
0.417	8.0	5936	0.4582	10.9131
0.4127	9.0	6678	0.4567	11.0361
0.4021	10.0	7420	0.4571	11.0566

Framework versions

• Transformers 4.48.1
• Pytorch 2.5.1+cu124
• Datasets 3.2.0
• Tokenizers 0.21.0