--- library_name: transformers license: apache-2.0 base_model: Helsinki-NLP/opus-mt-es-es tags: - generated_from_trainer model-index: - name: 8_6kmslsamples results: [] --- # 8_6kmslsamples This model is a fine-tuned version of [Helsinki-NLP/opus-mt-es-es](https://huggingface.co./Helsinki-NLP/opus-mt-es-es) on the None dataset. It achieves the following results on the evaluation set: - Loss: 0.4729 - Bleu Msl: 77.3631 - Bleu Asl: 0 - Ter Msl: 11.9346 - 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 Asl | Ter Msl | Ter Asl | |:-------------:|:-----:|:----:|:---------------:|:--------:|:--------:|:-------:|:-------:| | No log | 1.0 | 217 | 0.2646 | 21.2021 | 84.6183 | 57.1279 | 8.2427 | | No log | 2.0 | 434 | 0.1929 | 17.6495 | 84.9583 | 34.5912 | 7.4059 | | 0.4655 | 3.0 | 651 | 0.1742 | 67.4481 | 89.3963 | 19.9161 | 5.7531 | | 0.4655 | 4.0 | 868 | 0.1567 | 63.4679 | 86.5311 | 24.6855 | 7.8243 | | 0.0791 | 5.0 | 1085 | 0.1559 | 63.5398 | 82.6386 | 21.5933 | 9.6234 | | 0.0791 | 6.0 | 1302 | 0.1548 | 73.9641 | 86.8345 | 17.8721 | 8.1172 | | 0.0421 | 7.0 | 1519 | 0.1606 | 70.5606 | 88.7113 | 16.7715 | 5.8368 | | 0.0421 | 8.0 | 1736 | 0.1556 | 35.3257 | 86.5104 | 30.7652 | 7.4686 | | 0.0421 | 9.0 | 1953 | 0.1564 | 66.0661 | 85.8235 | 22.6415 | 8.2636 | | 0.0269 | 10.0 | 2170 | 0.1584 | 68.5372 | 88.0956 | 20.1782 | 6.7364 | | 0.0269 | 11.0 | 2387 | 0.1641 | 64.1749 | 89.0674 | 22.8512 | 5.8996 | | 0.0178 | 12.0 | 2604 | 0.1668 | 72.4073 | 88.8679 | 16.5094 | 6.0460 | | 0.0178 | 13.0 | 2821 | 0.1663 | 63.9882 | 86.2327 | 18.7107 | 8.8285 | | 0.0136 | 14.0 | 3038 | 0.1646 | 73.1200 | 88.9404 | 17.1908 | 6.2343 | | 0.0136 | 15.0 | 3255 | 0.1726 | 66.6138 | 87.2488 | 18.6583 | 6.9874 | | 0.0136 | 16.0 | 3472 | 0.1716 | 54.0153 | 87.6643 | 22.4843 | 6.4644 | | 0.0101 | 17.0 | 3689 | 0.1680 | 65.4861 | 88.5112 | 16.5618 | 5.7531 | | 0.0101 | 18.0 | 3906 | 0.1733 | 72.5027 | 88.9905 | 17.1384 | 5.8787 | | 0.0078 | 19.0 | 4123 | 0.1701 | 69.9889 | 85.8965 | 18.7631 | 8.7029 | | 0.0078 | 20.0 | 4340 | 0.1775 | 69.8437 | 88.9591 | 17.6101 | 6.1297 | | 0.0051 | 21.0 | 4557 | 0.1731 | 72.1034 | 88.3815 | 18.0818 | 6.5900 | | 0.0051 | 22.0 | 4774 | 0.1744 | 69.9773 | 89.2154 | 17.0860 | 5.7531 | | 0.0051 | 23.0 | 4991 | 0.1711 | 72.3725 | 90.1699 | 15.6709 | 5.3975 | | 0.0059 | 24.0 | 5208 | 0.1715 | 69.6558 | 89.0297 | 17.4004 | 6.1297 | | 0.0059 | 25.0 | 5425 | 0.1710 | 69.6504 | 87.9393 | 17.6101 | 7.2594 | | 0.0041 | 26.0 | 5642 | 0.1730 | 62.5723 | 87.0114 | 18.2914 | 8.3473 | | 0.0041 | 27.0 | 5859 | 0.1727 | 65.0893 | 88.5422 | 17.0860 | 6.9456 | | 0.0031 | 28.0 | 6076 | 0.1725 | 70.9388 | 88.1427 | 16.7191 | 7.3222 | | 0.0031 | 29.0 | 6293 | 0.1705 | 70.5254 | 88.4779 | 16.8239 | 6.9456 | | 0.0027 | 30.0 | 6510 | 0.1707 | 71.1488 | 88.5933 | 16.7715 | 6.9038 | ### Framework versions - Transformers 4.49.0 - Pytorch 2.5.1+cu124 - Datasets 3.3.2 - Tokenizers 0.21.0