467ad1a8dfb24caee6b2c28ddb48047e

This model is a fine-tuned version of Helsinki-NLP/opus-mt-tc-bible-big-deu_eng_fra_por_spa-mul on the Helsinki-NLP/opus_books [fr-nl] dataset. It achieves the following results on the evaluation set:

• Loss: 1.3539
• Data Size: 1.0
• Epoch Runtime: 62.3656
• Bleu: 8.1948

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	7.1463	0	5.5583	0.0995
No log	1	1000	3.5245	0.0078	5.9371	1.1871
No log	2	2000	2.7843	0.0156	6.3704	2.5197
No log	3	3000	2.3465	0.0312	7.8593	3.8312
0.0954	4	4000	2.0408	0.0625	9.5842	4.6602
1.9945	5	5000	1.8079	0.125	13.0710	5.4975
0.1064	6	6000	1.6156	0.25	20.5854	6.3412
0.1331	7	7000	1.4631	0.5	34.6998	6.9294
1.3124	8.0	8000	1.3311	1.0	63.1764	7.6202
1.1572	9.0	9000	1.2877	1.0	62.0938	7.9482
1.0234	10.0	10000	1.2728	1.0	63.1995	8.0784
0.9217	11.0	11000	1.2734	1.0	64.5127	8.0976
0.8327	12.0	12000	1.2947	1.0	63.6552	8.1607
0.7572	13.0	13000	1.3226	1.0	63.9981	8.1808
0.667	14.0	14000	1.3539	1.0	62.3656	8.1948

Framework versions

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