19b9a99d360126bde69d42d263b160bc

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

• Loss: 1.1429
• Data Size: 1.0
• Epoch Runtime: 1236.7036
• Bleu: 12.7860

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	2.6671	0	91.8338	1.6195
No log	1	2336	1.8218	0.0078	103.7495	8.9274
0.035	2	4672	1.7277	0.0156	115.6876	9.5350
0.0481	3	7008	1.6551	0.0312	136.6385	8.6207
1.9382	4	9344	1.5801	0.0625	171.9000	8.5264
1.8127	5	11680	1.5049	0.125	247.4238	8.6256
1.6469	6	14016	1.4036	0.25	390.1462	9.5051
1.5114	7	16352	1.3035	0.5	673.6065	10.3408
1.3464	8.0	18688	1.2034	1.0	1248.7078	11.2180
1.1972	9.0	21024	1.1447	1.0	1252.9235	11.8164
1.0886	10.0	23360	1.1153	1.0	1248.8658	12.3940
1.0169	11.0	25696	1.1019	1.0	1245.3960	12.6838
0.923	12.0	28032	1.0903	1.0	1236.4594	13.1384
0.8471	13.0	30368	1.0974	1.0	1232.4599	12.8015
0.8011	14.0	32704	1.1090	1.0	1231.2274	13.0162
0.7224	15.0	35040	1.1248	1.0	1236.4871	12.8928
0.6932	16.0	37376	1.1429	1.0	1236.7036	12.7860

Framework versions

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