4ae4f916c5b999108c177770a9ff594b

This model is a fine-tuned version of Helsinki-NLP/opus-mt-en-sv on the Helsinki-NLP/opus_books [fi-fr] dataset. It achieves the following results on the evaluation set:

• Loss: 3.0319
• Data Size: 1.0
• Epoch Runtime: 6.5354
• Bleu: 1.1434

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	7.6251	0	1.0672	0.0859
No log	1	88	7.0106	0.0078	1.4320	0.1071
No log	2	176	6.5327	0.0156	1.4227	0.1108
No log	3	264	6.2057	0.0312	1.5169	0.0805
No log	4	352	5.7218	0.0625	1.7557	0.0758
No log	5	440	5.1671	0.125	2.1989	0.1836
0.4257	6	528	4.5742	0.25	2.8416	0.2839
1.5822	7	616	4.0193	0.5	4.2644	0.4972
3.805	8.0	704	3.5764	1.0	6.5241	0.5910
3.5426	9.0	792	3.3554	1.0	6.2719	0.7215
3.2479	10.0	880	3.2131	1.0	6.3472	0.7900
3.0395	11.0	968	3.1192	1.0	6.2834	0.8791
2.869	12.0	1056	3.0514	1.0	6.4285	0.9258
2.7372	13.0	1144	3.0024	1.0	6.2123	0.9537
2.5937	14.0	1232	2.9745	1.0	6.2624	0.9970
2.4898	15.0	1320	2.9554	1.0	6.2847	1.0270
2.3743	16.0	1408	2.9432	1.0	6.1891	1.0857
2.2615	17.0	1496	2.9386	1.0	6.2949	1.0551
2.1486	18.0	1584	2.9418	1.0	6.7634	1.0560
2.0726	19.0	1672	2.9377	1.0	6.1594	1.1111
1.9606	20.0	1760	2.9539	1.0	6.2354	1.0555
1.8761	21.0	1848	2.9765	1.0	6.4182	1.0478
1.7622	22.0	1936	2.9921	1.0	6.4570	1.0483
1.677	23.0	2024	3.0319	1.0	6.5354	1.1434

Framework versions

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