40f1da89e7e824e2063f88a9511d41db

This model is a fine-tuned version of google-t5/t5-3b on the Helsinki-NLP/opus_books [es-nl] dataset. It achieves the following results on the evaluation set:

• Loss: 1.2136
• Data Size: 1.0
• Epoch Runtime: 369.3050
• Bleu: 6.8627

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	4.2405	0	23.5502	0.8837
No log	1	806	2.5116	0.0078	28.9944	2.4844
No log	2	1612	2.2724	0.0156	36.5960	3.4070
No log	3	2418	2.0877	0.0312	44.6501	3.8896
0.0784	4	3224	1.9084	0.0625	55.7682	3.9749
2.0165	5	4030	1.7600	0.125	77.1989	3.8466
1.8172	6	4836	1.6227	0.25	120.6444	4.5368
1.6367	7	5642	1.4664	0.5	201.3554	5.0837
1.4456	8.0	6448	1.3120	1.0	376.5444	6.2605
1.2701	9.0	7254	1.2362	1.0	377.9522	6.3447
1.1559	10.0	8060	1.1966	1.0	380.4398	6.6394
1.0685	11.0	8866	1.1744	1.0	374.9842	6.7114
0.9744	12.0	9672	1.1631	1.0	372.3183	6.8823
0.8847	13.0	10478	1.1593	1.0	381.9473	6.9412
0.8289	14.0	11284	1.1601	1.0	375.3103	6.7605
0.754	15.0	12090	1.1621	1.0	369.7971	6.9280
0.7096	16.0	12896	1.1852	1.0	374.5171	6.9066
0.6488	17.0	13702	1.2136	1.0	369.3050	6.8627

Framework versions

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