11661

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: 0.4527
• Model Preparation Time: 0.0032
• Bleu Msl: 0.0
• Bleu 1 Msl: 0.6915
• Bleu 2 Msl: 0.0153
• Bleu 3 Msl: 0.0046
• 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	Model Preparation Time	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	292	0.1642	0.0032	100.0000	0.5547	0.0102	0.0028	0.0014	100	100.0000	0.5429	0.0055	0.0013	0.0006	100
0.1267	2.0	584	0.1476	0.0032	100.0000	0.5937	0.0105	0.0029	0.0014	100	100.0000	0.5351	0.0055	0.0013	0.0006	100
0.1267	3.0	876	0.1483	0.0032	100.0000	0.6122	0.0107	0.0029	0.0014	100	100.0000	0.5390	0.0055	0.0013	0.0006	100
0.063	4.0	1168	0.1353	0.0032	100.0000	0.5362	0.0100	0.0028	0.0014	100	100.0000	0.5624	0.0056	0.0013	0.0006	100
0.063	5.0	1460	0.1427	0.0032	100.0000	0.5993	0.0106	0.0029	0.0014	100	100.0000	0.5507	0.0055	0.0013	0.0006	100
0.0391	6.0	1752	0.1448	0.0032	100.0000	0.4137	0.0088	0.0026	0.0013	100	100.0000	0.5318	0.0054	0.0013	0.0006	100
0.0266	7.0	2044	0.1471	0.0032	100.0000	0.4898	0.0095	0.0027	0.0013	100	100.0000	0.5446	0.0055	0.0013	0.0006	100
0.0266	8.0	2336	0.1447	0.0032	100.0000	0.5659	0.0103	0.0029	0.0014	100	100.0000	0.5457	0.0055	0.0013	0.0006	100
0.0234	9.0	2628	0.1435	0.0032	100.0000	0.6289	0.0108	0.0030	0.0014	100	100.0000	0.5702	0.0056	0.0013	0.0006	100
0.0234	10.0	2920	0.1389	0.0032	100.0000	0.6308	0.0108	0.0030	0.0014	100	100.0000	0.5624	0.0056	0.0013	0.0006	100
0.0162	11.0	3212	0.1413	0.0032	100.0000	0.5881	0.0105	0.0029	0.0014	100	100.0000	0.5708	0.0056	0.0013	0.0006	100
0.0138	12.0	3504	0.1458	0.0032	100.0000	0.6215	0.0107	0.0029	0.0014	100	100.0000	0.5797	0.0057	0.0013	0.0006	100
0.0138	13.0	3796	0.1439	0.0032	100.0000	0.5250	0.0099	0.0028	0.0014	100	100.0000	0.5585	0.0056	0.0013	0.0006	100
0.0105	14.0	4088	0.1482	0.0032	100.0000	0.5325	0.0099	0.0028	0.0014	100	100.0000	0.5569	0.0056	0.0013	0.0006	100
0.0105	15.0	4380	0.1524	0.0032	100.0000	0.4657	0.0093	0.0027	0.0013	100	100.0000	0.5468	0.0055	0.0013	0.0006	100
0.0098	16.0	4672	0.1519	0.0032	100.0000	0.5121	0.0098	0.0028	0.0013	100	100.0000	0.5535	0.0056	0.0013	0.0006	100
0.0098	17.0	4964	0.1569	0.0032	100.0000	0.4416	0.0091	0.0026	0.0013	100	100.0000	0.5557	0.0056	0.0013	0.0006	100
0.0081	18.0	5256	0.1524	0.0032	100.0000	0.5028	0.0097	0.0028	0.0013	100	100.0000	0.5474	0.0055	0.0013	0.0006	100
0.0062	19.0	5548	0.1493	0.0032	100.0000	0.4935	0.0096	0.0027	0.0013	100	100.0000	0.5552	0.0056	0.0013	0.0006	100
0.0062	20.0	5840	0.1513	0.0032	100.0000	0.5566	0.0102	0.0028	0.0014	100	100.0000	0.5452	0.0055	0.0013	0.0006	100
0.0053	21.0	6132	0.1471	0.0032	100.0000	0.5380	0.0100	0.0028	0.0014	100	100.0000	0.5619	0.0056	0.0013	0.0006	100
0.0053	22.0	6424	0.1480	0.0032	100.0000	0.5288	0.0099	0.0028	0.0014	100	100.0000	0.5541	0.0056	0.0013	0.0006	100
0.0042	23.0	6716	0.1489	0.0032	100.0000	0.5473	0.0101	0.0028	0.0014	100	100.0000	0.5641	0.0056	0.0013	0.0006	100
0.0039	24.0	7008	0.1500	0.0032	100.0000	0.6327	0.0108	0.0030	0.0014	100	100.0000	0.5602	0.0056	0.0013	0.0006	100
0.0039	25.0	7300	0.1496	0.0032	100.0000	0.5900	0.0105	0.0029	0.0014	100	100.0000	0.5624	0.0056	0.0013	0.0006	100
0.0032	26.0	7592	0.1478	0.0032	100.0000	0.5659	0.0103	0.0029	0.0014	100	100.0000	0.5630	0.0056	0.0013	0.0006	100
0.0032	27.0	7884	0.1493	0.0032	100.0000	0.5455	0.0101	0.0028	0.0014	100	100.0000	0.5647	0.0056	0.0013	0.0006	100
0.0031	28.0	8176	0.1500	0.0032	100.0000	0.5455	0.0101	0.0028	0.0014	100	100.0000	0.5674	0.0056	0.0013	0.0006	100
0.0031	29.0	8468	0.1504	0.0032	100.0000	0.5417	0.0100	0.0028	0.0014	100	100.0000	0.5630	0.0056	0.0013	0.0006	100
0.0021	30.0	8760	0.1502	0.0032	100.0000	0.5399	0.0100	0.0028	0.0014	100	100.0000	0.5635	0.0056	0.0013	0.0006	100

Framework versions

• Transformers 4.49.0
• Pytorch 2.5.1+cu124
• Datasets 3.3.2
• Tokenizers 0.21.0