1e586e9ff32b2bcecb6981b5521393de

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

• Loss: 1.9104
• Data Size: 1.0
• Epoch Runtime: 11.4654
• Bleu: 6.1340

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	3.9556	0	1.5061	1.2157
No log	1	33	3.8813	0.0078	2.1390	1.2060
No log	2	66	3.6950	0.0156	2.4420	1.3891
No log	3	99	3.3773	0.0312	2.8176	1.6731
0.2062	4	132	3.1119	0.0625	3.2074	2.1515
0.2062	5	165	2.7978	0.125	3.9415	2.7078
0.2062	6	198	2.6023	0.25	5.1174	2.9230
0.5676	7	231	2.4273	0.5	6.7223	3.1419
1.8254	8.0	264	2.2880	1.0	10.7664	3.8167
1.8254	9.0	297	2.2084	1.0	9.9693	3.8087
2.4664	10.0	330	2.1542	1.0	11.7658	3.9745
2.3094	11.0	363	2.1089	1.0	11.4597	4.3164
2.3094	12.0	396	2.0647	1.0	11.4472	4.6999
2.1714	13.0	429	2.0418	1.0	10.9065	5.0535
2.0846	14.0	462	2.0125	1.0	11.1601	5.2514
2.0846	15.0	495	2.0002	1.0	10.6700	5.2658
1.9877	16.0	528	1.9711	1.0	10.9546	5.3679
1.9197	17.0	561	1.9652	1.0	10.9433	5.3723
1.9197	18.0	594	1.9530	1.0	10.8528	5.5700
1.856	19.0	627	1.9373	1.0	10.5323	5.5289
1.7827	20.0	660	1.9305	1.0	11.5157	5.4426
1.7827	21.0	693	1.9219	1.0	11.4878	5.6287
1.7466	22.0	726	1.9147	1.0	12.0386	5.6536
1.6975	23.0	759	1.9140	1.0	11.6578	5.6462
1.6975	24.0	792	1.9145	1.0	11.4514	5.7523
1.6357	25.0	825	1.9036	1.0	11.1317	5.7489
1.5838	26.0	858	1.9055	1.0	11.8284	5.6285
1.5838	27.0	891	1.8995	1.0	11.7009	5.6704
1.5525	28.0	924	1.8983	1.0	11.3812	5.7482
1.4982	29.0	957	1.8979	1.0	12.0257	5.7826
1.4982	30.0	990	1.8926	1.0	11.6995	5.6236
1.4662	31.0	1023	1.9043	1.0	11.4540	5.8593
1.4408	32.0	1056	1.8969	1.0	11.4213	5.8493
1.4408	33.0	1089	1.8936	1.0	11.3773	5.9761
1.3954	34.0	1122	1.9104	1.0	11.4654	6.1340

Framework versions

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