2baf0c7fc011b0a22c6984d13a635a31

This model is a fine-tuned version of facebook/mbart-large-50-many-to-one-mmt on the Helsinki-NLP/opus_books [en-es] dataset. It achieves the following results on the evaluation set:

• Loss: 2.2368
• Data Size: 1.0
• Epoch Runtime: 604.6076
• Bleu: 9.1969

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	5.8102	0	48.1224	0.6052
No log	1	2336	4.1491	0.0078	52.6214	1.5574
0.0627	2	4672	3.6285	0.0156	58.7082	2.6285
0.0805	3	7008	3.2199	0.0312	67.5729	3.4881
2.9332	4	9344	2.8721	0.0625	83.5089	4.6759
2.6498	5	11680	2.5635	0.125	117.1398	5.4796
2.2932	6	14016	2.2973	0.25	190.0217	6.9013
2.0659	7	16352	2.0638	0.5	329.6237	7.8104
1.8034	8.0	18688	1.9001	1.0	607.7954	8.3532
1.523	9.0	21024	1.8715	1.0	608.0761	9.7238
1.3197	10.0	23360	1.8835	1.0	608.2041	9.1238
1.1199	11.0	25696	1.9745	1.0	603.7367	8.8339
0.9434	12.0	28032	2.0606	1.0	606.0338	9.1520
0.7766	13.0	30368	2.2368	1.0	604.6076	9.1969

Framework versions

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