628e87383e03fad5d721f1ba3f670dc0

This model is a fine-tuned version of google/mt5-xl on the Helsinki-NLP/opus_books [en-it] dataset. It achieves the following results on the evaluation set:

• Loss: 1.5476
• Data Size: 1.0
• Epoch Runtime: 383.8679
• Bleu: 8.1361

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	8.0100	0	25.0709	0.0149
No log	1	808	4.9109	0.0078	28.2408	0.0572
No log	2	1616	3.1439	0.0156	38.0719	0.3355
No log	3	2424	2.3399	0.0312	49.3007	0.6961
0.1226	4	3232	1.9302	0.0625	57.6411	4.3756
2.3361	5	4040	1.7976	0.125	81.3001	5.0607
2.0802	6	4848	1.7083	0.25	123.9069	5.8090
1.8453	7	5656	1.6199	0.5	211.6044	6.8138
1.7218	8.0	6464	1.5194	1.0	388.3820	7.3968
1.4987	9.0	7272	1.4779	1.0	385.1184	7.6968
1.36	10.0	8080	1.4696	1.0	381.5096	7.8967
1.2438	11.0	8888	1.4716	1.0	378.5281	7.9987
1.1021	12.0	9696	1.4820	1.0	381.7068	8.0585
1.033	13.0	10504	1.5100	1.0	381.6899	8.0742
0.9266	14.0	11312	1.5476	1.0	383.8679	8.1361

Framework versions

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