0ad935661942097dd5c0ae2b11a1e9af

This model is a fine-tuned version of google/mt5-xl on the Helsinki-NLP/opus_books [es-nl] dataset. It achieves the following results on the evaluation set:

• Loss: 1.6147
• Data Size: 1.0
• Epoch Runtime: 378.5897
• Bleu: 9.3592

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	8.8591	0	24.9839	0.0131
No log	1	806	3.7078	0.0078	27.9185	0.2891
No log	2	1612	2.8386	0.0156	37.8960	0.5919
No log	3	2418	2.2908	0.0312	45.4822	2.0559
0.114	4	3224	2.0603	0.0625	54.6514	4.8928
2.4355	5	4030	1.9097	0.125	79.3793	5.8108
2.1608	6	4836	1.7858	0.25	120.4500	6.6500
1.9512	7	5642	1.6737	0.5	209.1963	7.4743
1.7554	8.0	6448	1.5670	1.0	382.4961	8.3636
1.5612	9.0	7254	1.5219	1.0	376.1845	8.8128
1.4129	10.0	8060	1.5020	1.0	388.4327	9.0946
1.3086	11.0	8866	1.4992	1.0	377.5489	9.2108
1.1854	12.0	9672	1.5185	1.0	379.1931	9.3264
1.0602	13.0	10478	1.5451	1.0	377.0019	9.3584
0.9595	14.0	11284	1.5719	1.0	380.4620	9.4065
0.86	15.0	12090	1.6147	1.0	378.5897	9.3592

Framework versions

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