37ea563a4b411802628b7a9685742999

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

• Loss: 1.1325
• Data Size: 1.0
• Epoch Runtime: 1712.3020
• Bleu: 15.1463

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	3.0602	0	125.5759	2.6255
No log	1	3177	1.9635	0.0078	139.2434	7.3823
0.0392	2	6354	1.8381	0.0156	157.7000	7.0008
2.1836	3	9531	1.7466	0.0312	185.7243	7.2263
2.0194	4	12708	1.6474	0.0625	236.6189	7.6981
1.8331	5	15885	1.5397	0.125	349.4515	8.6908
1.6638	6	19062	1.4207	0.25	557.0253	9.3147
1.4703	7	22239	1.2966	0.5	971.5672	10.5970
1.2794	8.0	25416	1.1660	1.0	1812.9271	12.1729
1.1322	9.0	28593	1.1070	1.0	1813.4955	13.2869
1.0384	10.0	31770	1.0729	1.0	1812.9154	14.1803
0.9322	11.0	34947	1.0593	1.0	1808.4113	14.5935
0.8303	12.0	38124	1.0585	1.0	1804.2873	14.8567
0.7411	13.0	41301	1.0642	1.0	1808.3666	15.2248
0.6721	14.0	44478	1.0876	1.0	1725.7717	15.0559
0.6249	15.0	47655	1.1103	1.0	1709.6486	15.2645
0.5475	16.0	50832	1.1325	1.0	1712.3020	15.1463

Framework versions

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