9ab7d0181f4e205037241c02d5982c5f

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

• Loss: 1.3485
• Data Size: 1.0
• Epoch Runtime: 32.7055
• Bleu: 8.4854

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	2.5376	0	1.7160	1.8967
No log	1	31	2.2370	0.0078	2.1598	2.9674
No log	2	62	2.0193	0.0156	5.5509	4.7815
No log	3	93	1.9344	0.0312	10.8976	6.1564
No log	4	124	1.8286	0.0625	15.3905	5.8153
No log	5	155	1.6905	0.125	21.3750	5.5263
No log	6	186	1.5448	0.25	27.1492	5.4005
0.2929	7	217	1.4560	0.5	27.8635	6.4855
0.2929	8.0	248	1.3581	1.0	31.6103	9.0849
1.0376	9.0	279	1.3083	1.0	31.6960	8.3344
1.2251	10.0	310	1.3003	1.0	34.0091	8.7518
1.2251	11.0	341	1.3038	1.0	36.4410	9.2388
1.0437	12.0	372	1.3086	1.0	27.4977	7.9953
0.8906	13.0	403	1.3381	1.0	30.2501	8.1192
0.8906	14.0	434	1.3485	1.0	32.7055	8.4854

Framework versions

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