m2m100_412M_test

This model is a fine-tuned version of facebook/m2m100_418M on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.4156
• Bleu: 41.4971

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
3.9284	0.3615	500	0.5066	36.3923
0.4939	0.7231	1000	0.4463	38.6751
0.4483	1.0846	1500	0.4338	39.4755
0.4092	1.4461	2000	0.4258	40.2185
0.4099	1.8077	2500	0.4220	40.4820
0.3924	2.1692	3000	0.4192	41.1465
0.3723	2.5307	3500	0.4162	41.3847
0.3745	2.8923	4000	0.4156	41.4971

Framework versions

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