511b8646f4b981efd9878aad89ee57fe

This model is a fine-tuned version of google/mt5-xl on the Helsinki-NLP/opus_books [en-no] dataset. It achieves the following results on the evaluation set:

• Loss: 1.7171
• Data Size: 1.0
• Epoch Runtime: 52.3292
• Bleu: 11.0662

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	7.1882	0	3.3449	0.0340
No log	1	87	5.9851	0.0078	4.3107	0.0624
No log	2	174	4.3817	0.0156	11.2955	0.3101
No log	3	261	3.6612	0.0312	16.5874	0.6303
No log	4	348	3.1332	0.0625	25.2307	1.0858
0.2286	5	435	2.6152	0.125	20.7225	1.2132
0.744	6	522	2.0051	0.25	27.2097	9.0509
0.8299	7	609	1.7410	0.5	35.8400	8.9676
1.1673	8.0	696	1.6350	1.0	53.6893	9.6912
1.6693	9.0	783	1.6041	1.0	55.4095	10.1864
1.47	10.0	870	1.6003	1.0	49.4600	10.6361
1.2908	11.0	957	1.6075	1.0	50.4555	10.6224
1.1459	12.0	1044	1.6563	1.0	54.8101	10.7380
0.9882	13.0	1131	1.6470	1.0	49.3363	11.0055
0.8798	14.0	1218	1.7171	1.0	52.3292	11.0662

Framework versions

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