97444bf583dc91da2825bf394093e2ef

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

• Loss: 1.6102
• Data Size: 1.0
• Epoch Runtime: 203.4241
• Bleu: 9.5954

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	7.1472	0	14.8909	0.1248
No log	1	3177	3.7648	0.0078	16.9746	2.8094
0.0619	2	6354	3.1596	0.0156	18.5586	3.5432
2.9509	3	9531	2.7426	0.0312	22.0784	4.3965
2.5886	4	12708	2.4081	0.0625	27.7702	5.5281
2.2466	5	15885	2.1511	0.125	40.6274	6.5732
1.9989	6	19062	1.9294	0.25	63.0918	7.4797
1.7711	7	22239	1.7587	0.5	108.7908	8.2878
1.6023	8.0	25416	1.6072	1.0	205.7326	8.8817
1.422	9.0	28593	1.5576	1.0	201.3630	9.2482
1.3274	10.0	31770	1.5281	1.0	200.4891	9.5141
1.2075	11.0	34947	1.5227	1.0	201.4360	9.5176
1.0875	12.0	38124	1.5366	1.0	200.7618	9.5911
0.9913	13.0	41301	1.5502	1.0	203.1152	9.6452
0.921	14.0	44478	1.5788	1.0	199.5311	9.6271
0.865	15.0	47655	1.6102	1.0	203.4241	9.5954

Framework versions

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