df9829ef20751e04c89354d87ebedc3c

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

• Loss: 1.4839
• Data Size: 1.0
• Epoch Runtime: 938.8661
• Bleu: 10.6596

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	22.4540	0	68.4371	0.0104
No log	1	2336	14.3176	0.0078	76.8711	0.0187
0.2645	2	4672	6.1788	0.0156	85.6937	0.0188
0.1865	3	7008	2.5392	0.0312	101.7397	0.6592
2.6783	4	9344	2.0679	0.0625	130.2453	13.1640
2.3987	5	11680	1.9132	0.125	183.5239	7.1837
2.1995	6	14016	1.8011	0.25	287.3517	8.7471
2.0561	7	16352	1.7079	0.5	498.6651	9.5675
1.89	8.0	18688	1.6175	1.0	934.1045	9.6194
1.7455	9.0	21024	1.5634	1.0	963.7514	9.4783
1.6644	10.0	23360	1.5333	1.0	929.4754	9.9049
1.6017	11.0	25696	1.5107	1.0	934.2945	10.1029
1.5173	12.0	28032	1.4891	1.0	936.1568	10.4807
1.4493	13.0	30368	1.4805	1.0	930.4732	10.1182
1.4332	14.0	32704	1.4697	1.0	931.5481	10.3341
1.3594	15.0	35040	1.4615	1.0	928.7710	10.5791
1.3617	16.0	37376	1.4627	1.0	932.4758	10.5426
1.2551	17.0	39712	1.4575	1.0	930.8173	10.5483
1.2558	18.0	42048	1.4620	1.0	949.9990	10.8558
1.2071	19.0	44384	1.4663	1.0	942.6656	10.6449
1.1642	20.0	46720	1.4712	1.0	945.7716	10.5919
1.1119	21.0	49056	1.4839	1.0	938.8661	10.6596

Framework versions

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