89edc0df3a5b877e63e293bff90cb3bf

This model is a fine-tuned version of google/umt5-base on the Helsinki-NLP/opus_books [en-pt] dataset. It achieves the following results on the evaluation set:

• Loss: 11.8340
• Data Size: 0.125
• Epoch Runtime: 4.9392
• Bleu: 0.3225

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	11.9819	0	1.4201	0.3666
No log	1	35	11.7123	0.0078	1.7335	0.3419
No log	2	70	12.1181	0.0156	3.0176	0.2947
No log	3	105	12.1795	0.0312	4.0223	0.3097
No log	4	140	12.1218	0.0625	4.5280	0.2866
No log	5	175	11.8340	0.125	4.9392	0.3225

Framework versions

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