ramt-labse

This model is a fine-tuned version of mytranslatenisa/m2m100_en_ms_FineTuned_may on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.0333
• Bleu: 90.9007

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: 4
• eval_batch_size: 4
• seed: 42
• gradient_accumulation_steps: 8
• total_train_batch_size: 32
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 3
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Bleu
0.0448	0.9999	2630	0.0384	89.0090
0.0364	1.9997	5260	0.0351	87.0546
0.0257	2.9996	7890	0.0333	90.9007

Framework versions

• Transformers 4.44.0
• Pytorch 2.10.0+cu128
• Datasets 4.0.0
• Tokenizers 0.19.1