2fddfdad987dc50fd6cfaa79a290b942

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

• Loss: 1.2575
• Data Size: 1.0
• Epoch Runtime: 538.8969
• Bleu: 7.3004

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	3.2062	0	39.4263	0.4185
No log	1	966	2.5134	0.0078	43.2044	0.9992
No log	2	1932	2.3464	0.0156	51.3754	1.3163
0.0644	3	2898	2.1822	0.0312	62.9556	1.6390
2.5109	4	3864	2.0399	0.0625	78.6052	2.0907
2.3015	5	4830	1.9081	0.125	108.1160	2.6167
2.0713	6	5796	1.7503	0.25	168.7752	3.3277
1.8148	7	6762	1.5766	0.5	294.1465	4.1326
1.558	8.0	7728	1.4059	1.0	543.4787	5.1411
1.416	9.0	8694	1.3204	1.0	539.0476	5.7053
1.2935	10.0	9660	1.2631	1.0	539.2658	6.2669
1.1465	11.0	10626	1.2288	1.0	541.6563	6.5476
1.0686	12.0	11592	1.2230	1.0	542.1311	6.7940
0.9716	13.0	12558	1.2131	1.0	541.3803	6.9580
0.9213	14.0	13524	1.2145	1.0	541.7585	7.1673
0.8286	15.0	14490	1.2235	1.0	543.1369	7.2664
0.7748	16.0	15456	1.2399	1.0	538.1777	7.3070
0.7158	17.0	16422	1.2575	1.0	538.8969	7.3004

Framework versions

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