f91a23cecfbd289cf1348474d80a5c33

This model is a fine-tuned version of google/mt5-large on the Helsinki-NLP/opus_books [es-no] dataset. It achieves the following results on the evaluation set:

• Loss: 2.0487
• Data Size: 1.0
• Epoch Runtime: 42.6999
• Bleu: 5.9151

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
• distributed_type: multi-GPU
• num_devices: 4
• total_train_batch_size: 32
• total_eval_batch_size: 32
• optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: constant
• num_epochs: 50

Training results

Training Loss	Epoch	Step	Validation Loss	Data Size	Epoch Runtime	Bleu
No log	0	0	23.0375	0	3.6663	0.0086
No log	1	89	22.5653	0.0078	4.0129	0.0082
No log	2	178	18.5656	0.0156	7.1982	0.0092
No log	3	267	19.4656	0.0312	9.5094	0.0111
No log	4	356	14.8450	0.0625	11.7275	0.0136
No log	5	445	5.3013	0.125	14.7470	0.0381
0.9998	6	534	3.6382	0.25	17.6255	0.2923
1.8066	7	623	2.8759	0.5	27.2032	0.5935
3.1879	8.0	712	2.2714	1.0	44.2342	4.3190
2.6524	9.0	801	2.1324	1.0	44.1891	4.7504
2.512	10.0	890	2.0754	1.0	41.0263	5.1063
2.3432	11.0	979	2.0498	1.0	42.0423	5.2642
2.1826	12.0	1068	2.0215	1.0	41.9066	5.5056
2.0948	13.0	1157	2.0094	1.0	42.1578	5.5709
2.0081	14.0	1246	2.0083	1.0	43.5657	5.6278
1.929	15.0	1335	2.0098	1.0	42.0550	5.6109
1.8026	16.0	1424	2.0081	1.0	43.1516	5.8887
1.7762	17.0	1513	2.0119	1.0	41.7893	5.8135
1.6862	18.0	1602	2.0207	1.0	43.1757	5.8991
1.6062	19.0	1691	2.0286	1.0	42.1801	5.9735
1.546	20.0	1780	2.0487	1.0	42.6999	5.9151

Framework versions

• Transformers 4.57.0
• Pytorch 2.8.0+cu128
• Datasets 4.2.0
• Tokenizers 0.22.1