byt5-base-es_pbb

This model is a fine-tuned version of google/byt5-base on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.8368
• Bleu: 2.9665
• Gen Len: 100.0518

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: 16
• eval_batch_size: 16
• seed: 65
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 10.0

Training results

Training Loss	Epoch	Step	Validation Loss	Bleu	Gen Len
No log	1.0	397	1.1680	0.0083	19.0
1.7159	2.0	794	1.0287	0.0165	19.0
1.1916	3.0	1191	0.9632	0.0177	19.0
1.077	4.0	1588	0.9212	0.0307	19.0
1.077	5.0	1985	0.8921	0.0278	19.0
1.0159	6.0	2382	0.8694	0.0465	19.0
0.9723	7.0	2779	0.8543	0.0489	19.0
0.9457	8.0	3176	0.8452	0.0451	19.0
0.9276	9.0	3573	0.8381	0.0496	19.0
0.9276	10.0	3970	0.8368	0.048	19.0

Framework versions

• Transformers 4.27.4
• Pytorch 2.0.0+cu118
• Datasets 2.11.0
• Tokenizers 0.13.3