96aa4589038cb435dd1101e5d891e0ba

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

• Loss: 1.4117
• Data Size: 1.0
• Epoch Runtime: 47.9806
• Bleu: 9.4832

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	2.4446	0	3.4864	2.0941
No log	1	77	2.3261	0.0078	3.7524	2.2082
No log	2	154	2.1595	0.0156	8.4747	3.0719
No log	3	231	2.0865	0.0312	12.2365	3.8498
No log	4	308	2.0177	0.0625	17.0066	4.8276
No log	5	385	1.9267	0.125	23.1477	6.9463
0.1963	6	462	1.8028	0.25	26.4708	5.5171
0.8057	7	539	1.6651	0.5	31.9390	5.8217
1.6499	8.0	616	1.5084	1.0	54.4675	7.1576
1.477	9.0	693	1.4191	1.0	56.7229	7.7309
1.2294	10.0	770	1.3795	1.0	49.1534	8.3925
1.1492	11.0	847	1.3570	1.0	50.9544	8.7088
0.9643	12.0	924	1.3509	1.0	44.6494	8.8015
0.8548	13.0	1001	1.3603	1.0	47.7112	9.1627
0.7597	14.0	1078	1.3679	1.0	52.3373	9.4660
0.6788	15.0	1155	1.3886	1.0	47.0834	9.5313
0.6126	16.0	1232	1.4117	1.0	47.9806	9.4832

Framework versions

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