d867bf9130672c1a8b62bf9fc0b7f602

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-it] dataset. It achieves the following results on the evaluation set:

• Loss: 2.1582
• Data Size: 1.0
• Epoch Runtime: 24.4510
• Bleu: 4.7820

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	7.2676	0	2.3496	0.0663
No log	1	367	4.7986	0.0078	2.9380	0.8384
No log	2	734	3.9781	0.0156	3.0596	1.2546
No log	3	1101	3.4943	0.0312	4.1123	1.7892
No log	4	1468	3.0759	0.0625	4.9290	2.3078
0.1807	5	1835	2.7368	0.125	6.7183	2.8804
2.6883	6	2202	2.4612	0.25	9.3884	3.4075
2.3474	7	2569	2.2636	0.5	14.3512	3.8673
2.0452	8.0	2936	2.0837	1.0	26.8325	4.4197
1.7936	9.0	3303	2.0336	1.0	25.1930	4.5775
1.5745	10.0	3670	2.0215	1.0	25.1768	4.7023
1.4022	11.0	4037	2.0378	1.0	24.3836	4.7877
1.2812	12.0	4404	2.0513	1.0	26.1829	4.6533
1.164	13.0	4771	2.1157	1.0	25.3948	4.7247
0.9903	14.0	5138	2.1582	1.0	24.4510	4.7820

Framework versions

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