b5673f0ec8b577ee4b2991f2d4af8c5b

This model is a fine-tuned version of google/umt5-xl on the Helsinki-NLP/opus_books [es-no] dataset. It achieves the following results on the evaluation set:

• Loss: 2.1061
• Data Size: 1.0
• Epoch Runtime: 59.4084
• Bleu: 8.1508

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	5.8285	0	4.0429	1.7249
No log	1	89	5.5023	0.0078	4.4242	2.7088
No log	2	178	4.5237	0.0156	9.1380	5.5642
No log	3	267	4.1058	0.0312	17.9356	7.4000
No log	4	356	3.6054	0.0625	25.7074	9.2739
No log	5	445	3.0711	0.125	26.6073	13.0192
0.3122	6	534	2.5416	0.25	31.1763	15.5088
1.076	7	623	2.2106	0.5	45.7976	6.6268
2.3556	8.0	712	2.0000	1.0	66.0511	7.4405
2.0185	9.0	801	1.9424	1.0	64.9103	7.6023
1.7966	10.0	890	1.9292	1.0	58.2939	7.9839
1.5756	11.0	979	1.9661	1.0	65.0326	8.1567
1.3635	12.0	1068	2.0052	1.0	58.8789	8.1580
1.2366	13.0	1157	2.0370	1.0	58.4167	8.1711
1.09	14.0	1246	2.1061	1.0	59.4084	8.1508

Framework versions

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