8fd3457735fa689adafb606cb4072f06

This model is a fine-tuned version of google-t5/t5-3b on the Helsinki-NLP/opus_books [de-it] dataset. It achieves the following results on the evaluation set:

• Loss: 1.3508
• Data Size: 1.0
• Epoch Runtime: 333.2017
• Bleu: 8.3836

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	2.9101	0	21.8176	0.9591
No log	1	684	2.3145	0.0078	23.8740	2.4121
No log	2	1368	2.0432	0.0156	33.2745	5.0483
No log	3	2052	1.9144	0.0312	44.7816	7.1777
No log	4	2736	1.8023	0.0625	50.8609	6.2695
1.9542	5	3420	1.7062	0.125	67.7002	6.0525
1.8421	6	4104	1.6101	0.25	100.9292	6.8973
1.6355	7	4788	1.4894	0.5	176.6196	7.8111
1.4844	8.0	5472	1.3753	1.0	331.7755	7.9854
1.3137	9.0	6156	1.3177	1.0	316.6885	8.2918
1.1997	10.0	6840	1.2847	1.0	319.6870	8.3214
1.0816	11.0	7524	1.2786	1.0	325.7775	8.3826
0.9921	12.0	8208	1.2695	1.0	310.9514	8.3532
0.9047	13.0	8892	1.2836	1.0	326.1859	8.6761
0.8451	14.0	9576	1.2906	1.0	324.5127	8.3699
0.7594	15.0	10260	1.3217	1.0	330.3170	8.4491
0.699	16.0	10944	1.3508	1.0	333.2017	8.3836

Framework versions

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