792304578eeaaf421d91b32ca9e3f8d5

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

• Loss: 1.1469
• Data Size: 1.0
• Epoch Runtime: 641.1222
• Bleu: 14.9571

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	7.5319	0	42.1921	0.0227
No log	1	1407	2.4063	0.0078	47.0469	1.2099
No log	2	2814	1.7253	0.0156	59.5257	1.5937
0.0687	3	4221	1.4626	0.0312	73.6290	9.3270
1.7852	4	5628	1.3730	0.0625	94.4569	10.1598
1.6029	5	7035	1.2781	0.125	126.9140	11.3849
1.4152	6	8442	1.1953	0.25	199.0452	13.0616
1.3277	7	9849	1.1298	0.5	347.5435	13.6878
1.1487	8.0	11256	1.0761	1.0	644.9083	14.2348
1.0115	9.0	12663	1.0489	1.0	642.3715	14.7043
0.9556	10.0	14070	1.0474	1.0	639.1853	14.9049
0.8314	11.0	15477	1.0551	1.0	639.4692	14.8704
0.7299	12.0	16884	1.0761	1.0	641.4331	15.0293
0.6821	13.0	18291	1.1074	1.0	641.8712	14.9957
0.6183	14.0	19698	1.1469	1.0	641.1222	14.9571

Framework versions

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