byt5-base-es_kog

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: 0.8694
• Bleu: 1.878
• Gen Len: 100.4987

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	1.1707	0.0631	19.0
1.6442	2.0	794	1.0400	0.1214	19.0
1.1712	3.0	1191	0.9847	0.1063	19.0
1.0724	4.0	1588	0.9467	0.1209	19.0
1.0724	5.0	1985	0.9176	0.1059	19.0
1.016	6.0	2382	0.8977	0.1166	19.0
0.9799	7.0	2779	0.8852	0.1481	19.0
0.9553	8.0	3176	0.8748	0.1381	19.0
0.9386	9.0	3573	0.8716	0.1294	19.0
0.9386	10.0	3970	0.8694	0.1337	19.0

Framework versions

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