t5-small-translation-opus

This model is a fine-tuned version of google-t5/t5-small on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 1.5599
• Bleu: 6.6121
• Gen Len: 18.3408

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: 1e-05
• train_batch_size: 32
• eval_batch_size: 32
• seed: 42
• optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• num_epochs: 10
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Bleu	Gen Len
1.9309	1.0	3178	1.6953	5.5146	18.4035
1.8747	2.0	6356	1.6464	5.962	18.3844
1.8248	3.0	9534	1.6186	6.1886	18.3686
1.815	4.0	12712	1.5988	6.3108	18.358
1.8006	5.0	15890	1.5852	6.4011	18.3488
1.7824	6.0	19068	1.5758	6.5065	18.3438
1.7675	7.0	22246	1.5682	6.5584	18.3466
1.7546	8.0	25424	1.5636	6.5855	18.3418
1.7528	9.0	28602	1.5607	6.6009	18.3427
1.7572	10.0	31780	1.5599	6.6121	18.3408

Framework versions

• Transformers 4.55.2
• Pytorch 2.8.0+cu128
• Datasets 3.6.0
• Tokenizers 0.21.4