5508bea1106dcf8f38928c258e582fd3

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 [en-sv] dataset. It achieves the following results on the evaluation set:

• Loss: 1.8144
• Data Size: 1.0
• Epoch Runtime: 7.6010
• Bleu: 6.9931

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	7.4686	0	0.9839	0.0456
No log	1	77	6.1587	0.0078	1.3547	0.1921
No log	2	154	4.8660	0.0156	1.7285	0.7242
No log	3	231	3.9128	0.0312	2.4119	0.9697
No log	4	308	3.2300	0.0625	2.5775	1.6177
No log	5	385	2.6993	0.125	2.8676	2.8669
0.3029	6	462	2.2712	0.25	3.4343	4.1236
0.9005	7	539	1.9408	0.5	4.7194	5.4670
1.6634	8.0	616	1.7178	1.0	7.2565	6.8976
1.3766	9.0	693	1.6382	1.0	6.3484	6.8624
1.0033	10.0	770	1.6518	1.0	6.6554	6.9847
0.9036	11.0	847	1.7171	1.0	7.0013	7.0927
0.6668	12.0	924	1.7463	1.0	7.1039	7.0672
0.5555	13.0	1001	1.8144	1.0	7.6010	6.9931

Framework versions

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