2661

This model is a fine-tuned version of Helsinki-NLP/opus-mt-es-es on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.4495
• Bleu Msl: 0.0
• Bleu 1 Msl: 0.7593
• Bleu 2 Msl: 0.0161
• Bleu 3 Msl: 0.0047
• Bleu 4 Msl: 0.0023
• Ter Msl: 100
• Bleu Asl: 0
• Bleu 1 Asl: 0
• Bleu 2 Asl: 0
• Bleu 3 Asl: 0
• Bleu 4 Asl: 0
• Ter Asl: 100

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: 0.0001
• train_batch_size: 32
• eval_batch_size: 64
• seed: 42
• optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• num_epochs: 30
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Bleu Msl	Bleu 1 Msl	Bleu 2 Msl	Bleu 3 Msl	Bleu 4 Msl	Ter Msl	Bleu Asl	Bleu 1 Asl	Bleu 2 Asl	Bleu 3 Asl	Bleu 4 Asl	Ter Asl
No log	1.0	67	0.4200	42.7287	0.6604	0.0111	0.0030	0.0014	100	0	0	0	0	0	100
No log	2.0	134	0.3958	50.8133	0.7186	0.0116	0.0031	0.0015	100	0	0	0	0	0	100
No log	3.0	201	0.3869	70.7107	0.7317	0.0117	0.0031	0.0015	100	0	0	0	0	0	100
No log	4.0	268	0.3764	70.7107	0.7336	0.0117	0.0031	0.0015	100	0	0	0	0	0	100
No log	5.0	335	0.3818	50.8133	0.7261	0.0117	0.0031	0.0015	100	0	0	0	0	0	100
No log	6.0	402	0.3615	50.8133	0.7467	0.0118	0.0032	0.0015	100	0	0	0	0	0	100
No log	7.0	469	0.3790	50.8133	0.7336	0.0117	0.0031	0.0015	100	0	0	0	0	0	100
0.1217	8.0	536	0.3946	50.0000	0.7411	0.0118	0.0031	0.0015	100	0	0	0	0	0	100
0.1217	9.0	603	0.4057	45.1801	0.7448	0.0118	0.0032	0.0015	100	0	0	0	0	0	100
0.1217	10.0	670	0.3968	70.7107	0.7392	0.0118	0.0031	0.0015	100	0	0	0	0	0	100
0.1217	11.0	737	0.4003	70.7107	0.7411	0.0118	0.0031	0.0015	100	0	0	0	0	0	100
0.1217	12.0	804	0.3982	70.7107	0.7561	0.0119	0.0032	0.0015	100	0	0	0	0	0	100
0.1217	13.0	871	0.4016	70.7107	0.7486	0.0119	0.0032	0.0015	100	0	0	0	0	0	100
0.1217	14.0	938	0.4065	50.8133	0.7542	0.0119	0.0032	0.0015	100	0	0	0	0	0	100
0.0287	15.0	1005	0.3989	50.0000	0.7561	0.0119	0.0032	0.0015	100	0	0	0	0	0	100
0.0287	16.0	1072	0.4009	50.0000	0.7523	0.0119	0.0032	0.0015	100	0	0	0	0	0	100
0.0287	17.0	1139	0.4154	50.0000	0.7505	0.0119	0.0032	0.0015	100	0	0	0	0	0	100
0.0287	18.0	1206	0.4111	70.7107	0.7411	0.0118	0.0031	0.0015	100	0	0	0	0	0	100
0.0287	19.0	1273	0.4163	70.7107	0.7448	0.0118	0.0032	0.0015	100	0	0	0	0	0	100
0.0287	20.0	1340	0.4134	70.7107	0.7542	0.0119	0.0032	0.0015	100	0	0	0	0	0	100
0.0287	21.0	1407	0.4165	70.7107	0.7542	0.0119	0.0032	0.0015	100	0	0	0	0	0	100
0.0287	22.0	1474	0.4120	50.0000	0.7486	0.0119	0.0032	0.0015	100	0	0	0	0	0	100
0.0132	23.0	1541	0.4199	50.0000	0.7467	0.0118	0.0032	0.0015	100	0	0	0	0	0	100
0.0132	24.0	1608	0.4200	50.0000	0.7523	0.0119	0.0032	0.0015	100	0	0	0	0	0	100
0.0132	25.0	1675	0.4226	70.7107	0.7523	0.0119	0.0032	0.0015	100	0	0	0	0	0	100
0.0132	26.0	1742	0.4199	70.7107	0.7542	0.0119	0.0032	0.0015	100	0	0	0	0	0	100
0.0132	27.0	1809	0.4216	70.7107	0.7523	0.0119	0.0032	0.0015	100	0	0	0	0	0	100
0.0132	28.0	1876	0.4225	50.8133	0.7542	0.0119	0.0032	0.0015	100	0	0	0	0	0	100
0.0132	29.0	1943	0.4232	50.8133	0.7523	0.0119	0.0032	0.0015	100	0	0	0	0	0	100
0.0087	30.0	2010	0.4235	50.8133	0.7523	0.0119	0.0032	0.0015	100	0	0	0	0	0	100

Framework versions

• Transformers 4.48.3
• Pytorch 2.5.1+cu124
• Datasets 3.3.2
• Tokenizers 0.21.0