710d80b5ad433be146eac1c3f9f6b3a5

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

• Loss: 1.8863
• Data Size: 1.0
• Epoch Runtime: 54.7240
• Bleu: 6.9620

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	7.8905	0	3.4834	0.0179
No log	1	89	6.7845	0.0078	3.8369	0.0146
No log	2	178	4.4131	0.0156	11.0523	0.1645
No log	3	267	3.8186	0.0312	17.4287	0.3608
No log	4	356	3.3102	0.0625	22.5856	0.5910
No log	5	445	2.7230	0.125	21.9316	0.7283
0.3037	6	534	2.1537	0.25	26.6146	4.5924
0.9462	7	623	1.9278	0.5	41.2115	5.4022
2.1336	8.0	712	1.8363	1.0	62.8856	6.0140
1.853	9.0	801	1.7828	1.0	52.1380	6.5381
1.6556	10.0	890	1.7905	1.0	51.7727	6.6433
1.452	11.0	979	1.8083	1.0	55.2504	6.8516
1.2686	12.0	1068	1.8476	1.0	51.9475	7.0515
1.1406	13.0	1157	1.8863	1.0	54.7240	6.9620

Framework versions

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