2bd201832a6c0022b0c889002ad872ff

This model is a fine-tuned version of google/mt5-xl on the Helsinki-NLP/opus_books [de-nl] dataset. It achieves the following results on the evaluation set:

• Loss: 1.8220
• Data Size: 1.0
• Epoch Runtime: 185.5742
• Bleu: 9.8214

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	8.1471	0	12.2683	0.0289
No log	1	390	4.6064	0.0078	14.0348	0.1720
No log	2	780	3.6595	0.0156	19.6016	0.4727
No log	3	1170	2.9152	0.0312	28.0245	0.8404
No log	4	1560	2.3729	0.0625	35.4504	3.1282
0.204	5	1950	2.0826	0.125	46.1279	6.0931
0.3659	6	2340	1.9362	0.25	61.9895	7.1007
2.1779	7	2730	1.8420	0.5	105.0241	7.8619
1.9347	8.0	3120	1.7334	1.0	186.1186	8.6843
1.7092	9.0	3510	1.6959	1.0	182.5112	9.2084
1.5389	10.0	3900	1.6837	1.0	182.6115	9.4217
1.4014	11.0	4290	1.6824	1.0	186.1512	9.6612
1.2698	12.0	4680	1.7072	1.0	184.9993	9.8539
1.168	13.0	5070	1.7316	1.0	182.9180	9.8787
1.0595	14.0	5460	1.7575	1.0	183.1962	9.8212
0.9316	15.0	5850	1.8220	1.0	185.5742	9.8214

Framework versions

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