80b7f30682f88ece747372a14ab620bc

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

• Loss: 2.1627
• Data Size: 1.0
• Epoch Runtime: 24.2616
• Bleu: 1.6301

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	4.0631	0	2.4335	0.1246
No log	1	89	4.0072	0.0078	2.9322	0.1232
No log	2	178	3.7818	0.0156	3.3542	0.1304
No log	3	267	3.6394	0.0312	3.9202	0.1546
No log	4	356	3.5167	0.0625	5.2122	0.1518
No log	5	445	3.3866	0.125	6.8490	0.2068
0.2434	6	534	3.2375	0.25	8.4315	0.1871
1.1677	7	623	3.0671	0.5	12.9446	0.3095
3.116	8.0	712	2.8850	1.0	24.8634	0.4666
2.9525	9.0	801	2.7724	1.0	23.7692	0.5939
2.8776	10.0	890	2.6974	1.0	22.4813	0.6524
2.788	11.0	979	2.6302	1.0	22.7156	0.8093
2.6853	12.0	1068	2.5706	1.0	22.9638	0.8036
2.6173	13.0	1157	2.5230	1.0	23.2780	0.9382
2.5673	14.0	1246	2.4776	1.0	23.5736	0.9681
2.5152	15.0	1335	2.4404	1.0	22.8766	0.9690
2.4372	16.0	1424	2.4092	1.0	23.2768	1.0865
2.4121	17.0	1513	2.3803	1.0	25.0582	1.1240
2.3656	18.0	1602	2.3489	1.0	23.5105	1.1894
2.3101	19.0	1691	2.3271	1.0	23.9473	1.2188
2.2771	20.0	1780	2.3082	1.0	23.6225	1.1815
2.2253	21.0	1869	2.2885	1.0	23.9636	1.2601
2.194	22.0	1958	2.2772	1.0	23.8962	1.2873
2.1742	23.0	2047	2.2636	1.0	23.5343	1.3025
2.1198	24.0	2136	2.2420	1.0	23.2088	1.3515
2.0858	25.0	2225	2.2315	1.0	24.9648	1.3373
2.0456	26.0	2314	2.2180	1.0	24.1799	1.3876
2.0293	27.0	2403	2.2133	1.0	24.0982	1.4022
2.0236	28.0	2492	2.2002	1.0	24.7002	1.4988
1.9647	29.0	2581	2.1906	1.0	24.1223	1.4973
1.9522	30.0	2670	2.1843	1.0	23.9940	1.4922
1.9203	31.0	2759	2.1845	1.0	23.5057	1.4814
1.8914	32.0	2848	2.1737	1.0	25.4286	1.5828
1.8595	33.0	2937	2.1672	1.0	24.4033	1.6258
1.85	34.0	3026	2.1685	1.0	25.6600	1.5604
1.8111	35.0	3115	2.1683	1.0	24.5606	1.5346
1.7959	36.0	3204	2.1615	1.0	24.4040	1.5955
1.7818	37.0	3293	2.1587	1.0	23.9411	1.5674
1.7527	38.0	3382	2.1564	1.0	24.0609	1.5848
1.7276	39.0	3471	2.1523	1.0	24.0727	1.5702
1.7115	40.0	3560	2.1576	1.0	24.7035	1.6662
1.6809	41.0	3649	2.1516	1.0	26.0027	1.6384
1.657	42.0	3738	2.1534	1.0	24.9446	1.6151
1.6367	43.0	3827	2.1560	1.0	24.3666	1.6189
1.6194	44.0	3916	2.1518	1.0	24.4493	1.6453
1.6076	45.0	4005	2.1627	1.0	24.2616	1.6301

Framework versions

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