dfc6b9d82f6c528dd8093bf5cb8a3d4d

This model is a fine-tuned version of google-t5/t5-3b on the Helsinki-NLP/opus_books [en-no] dataset. It achieves the following results on the evaluation set:

• Loss: 1.5431
• Data Size: 1.0
• Epoch Runtime: 52.7750
• Bleu: 8.1375

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	2.7304	0	3.5466	1.5681
No log	1	87	2.4932	0.0078	3.9387	1.7255
No log	2	174	2.3011	0.0156	7.5761	2.3471
No log	3	261	2.2255	0.0312	12.6370	2.7873
No log	4	348	2.1260	0.0625	17.0902	3.4511
0.1001	5	435	2.0060	0.125	20.7682	4.0295
0.5146	6	522	1.8729	0.25	25.8551	4.3779
0.6839	7	609	1.7406	0.5	31.8658	5.2788
1.017	8.0	696	1.5912	1.0	56.5105	6.2743
1.4721	9.0	783	1.5157	1.0	50.1307	6.8718
1.2861	10.0	870	1.4703	1.0	53.3501	7.4851
1.1385	11.0	957	1.4616	1.0	48.4484	7.7546
1.0167	12.0	1044	1.4783	1.0	51.4189	7.9532
0.9032	13.0	1131	1.4647	1.0	48.0732	7.9947
0.8108	14.0	1218	1.5021	1.0	49.3293	8.2102
0.7275	15.0	1305	1.5431	1.0	52.7750	8.1375

Framework versions

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