ntviet/mbart-hre-viet1.0

mBART Hre Vietnamese translation 1.0

This model is a fine-tuned version of facebook/mbart-large-50-many-to-many-mmt on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.0011

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: 8
• eval_batch_size: 8
• 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: 20

Training results

Training Loss	Epoch	Step	Validation Loss
No log	1.0	336	0.1069
0.5938	2.0	672	0.0613
0.0798	3.0	1008	0.0315
0.0798	4.0	1344	0.0154
0.0267	5.0	1680	0.0098
0.0123	6.0	2016	0.0057
0.0123	7.0	2352	0.0062
0.0061	8.0	2688	0.0052
0.0043	9.0	3024	0.0033
0.0043	10.0	3360	0.0031
0.003	11.0	3696	0.0023
0.0024	12.0	4032	0.0023
0.0024	13.0	4368	0.0020
0.0018	14.0	4704	0.0016
0.0016	15.0	5040	0.0020
0.0016	16.0	5376	0.0015
0.0014	17.0	5712	0.0013
0.0011	18.0	6048	0.0011
0.0011	19.0	6384	0.0010
0.0008	20.0	6720	0.0011

Framework versions

• Transformers 4.47.1
• Pytorch 2.5.1+cu124
• Datasets 3.2.0
• Tokenizers 0.21.0