Helsinki-NLP-opus-mt-en-es-fine-tuned

This model is a fine-tuned version of Helsinki-NLP/opus-mt-en-es on the None dataset. It achieves the following results on the evaluation set:

• Loss: 1.9264
• Bleu: 59.0232
• Gen Len: 10.2241

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: 1e-05
• train_batch_size: 32
• eval_batch_size: 128
• seed: 42
• optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED 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: 1
• mixed_precision_training: Native AMP
• label_smoothing_factor: 0.1

Training results

Training Loss	Epoch	Step	Validation Loss	Bleu	Gen Len
2.0075	0.0571	100	1.9689	56.5918	10.0929
1.9751	0.1143	200	1.9573	57.4701	10.1606
1.9695	0.1714	300	1.9491	58.1358	10.1936
1.9642	0.2286	400	1.9454	58.2964	10.2016
1.9675	0.2857	500	1.9430	58.4552	10.1803
1.9665	0.3429	600	1.9397	58.774	10.1874
1.9698	0.4	700	1.9366	58.7652	10.2136
1.9567	0.4571	800	1.9351	58.9254	10.208
1.9647	0.5143	900	1.9332	58.971	10.2239
1.9562	0.5714	1000	1.9311	59.0491	10.2169
1.9476	0.6286	1100	1.9302	58.9443	10.2049
1.956	0.6857	1200	1.9284	59.0869	10.228
1.9479	0.7429	1300	1.9286	58.9766	10.22
1.9402	0.8	1400	1.9274	59.0948	10.2253
1.9384	0.8571	1500	1.9272	59.0634	10.224
1.9473	0.9143	1600	1.9269	59.062	10.2177
1.9416	0.9714	1700	1.9264	59.0232	10.2241

Framework versions

• Transformers 4.57.2
• Pytorch 2.9.0+cu126
• Datasets 4.0.0
• Tokenizers 0.22.1