3c38c0481a0037f746b5d0f4ebf541b9

This model is a fine-tuned version of google/long-t5-tglobal-xl on the Helsinki-NLP/opus_books [es-no] dataset. It achieves the following results on the evaluation set:

• Loss: 1.7950
• Data Size: 1.0
• Epoch Runtime: 61.8989
• Bleu: 2.0930

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	3.5779	0	4.4405	0.0548
No log	1	89	2.9932	0.0078	5.5066	0.0636
No log	2	178	2.6780	0.0156	10.4810	0.3789
No log	3	267	2.5460	0.0312	16.8449	0.5178
No log	4	356	2.4467	0.0625	23.9740	0.4745
No log	5	445	2.3447	0.125	28.8034	0.3841
0.1979	6	534	2.2146	0.25	31.4532	0.4716
0.8908	7	623	2.1019	0.5	43.7093	0.7499
2.2973	8.0	712	1.9769	1.0	71.7201	0.8643
2.0825	9.0	801	1.8884	1.0	60.5489	1.2267
1.9589	10.0	890	1.8386	1.0	61.2092	1.3377
1.8246	11.0	979	1.8128	1.0	62.9494	1.5261
1.7046	12.0	1068	1.7875	1.0	59.8757	1.6324
1.6049	13.0	1157	1.7712	1.0	62.5767	1.6917
1.5218	14.0	1246	1.7567	1.0	61.1905	1.7882
1.4302	15.0	1335	1.7595	1.0	61.4564	1.8567
1.3283	16.0	1424	1.7606	1.0	62.1704	1.9414
1.2638	17.0	1513	1.7693	1.0	67.3304	1.9924
1.1834	18.0	1602	1.7950	1.0	61.8989	2.0930

Framework versions

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