m2m100_418M_pt_formal

This model is a fine-tuned version of facebook/m2m100_418M on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.3145
• Bleu: 40.5054

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: 32
• 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
• lr_scheduler_warmup_steps: 500
• num_epochs: 3.0

Training results

Training Loss	Epoch	Step	Validation Loss	Bleu
4.0262	0.3956	500	0.3792	36.1929
0.3736	0.7911	1000	0.3307	38.8103
0.3346	1.1867	1500	0.3236	39.2764
0.3118	1.5823	2000	0.3194	39.7404
0.3084	1.9778	2500	0.3160	40.0558
0.2856	2.3734	3000	0.3154	40.2045
0.277	2.7690	3500	0.3145	40.5054

Framework versions

• Transformers 4.48.3
• Pytorch 2.5.1+cu124
• Datasets 3.3.2
• Tokenizers 0.21.0