3317d5970d95d309b5c8a566c613125c

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

• Loss: 12.8673
• Data Size: 0.125
• Epoch Runtime: 7.3437
• Bleu: 0.2699

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	12.6620	0	2.0593	0.2629
No log	1	70	12.6427	0.0078	2.3722	0.2385
No log	2	140	12.8476	0.0156	3.1829	0.2655
No log	3	210	13.1023	0.0312	4.2745	0.2462
No log	4	280	13.0093	0.0625	5.9938	0.2026
No log	5	350	12.8673	0.125	7.3437	0.2699

Framework versions

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