byt5_1k

This model was trained from scratch on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.0868

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: 0.0001
• train_batch_size: 400
• eval_batch_size: 800
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 20

Training results

Training Loss	Epoch	Step	Validation Loss
No log	1.0	3	0.1081
No log	2.0	6	0.0983
No log	3.0	9	0.1285
0.1432	4.0	12	0.0961
0.1432	5.0	15	0.1040
0.1432	6.0	18	0.1032
0.1488	7.0	21	0.0938
0.1488	8.0	24	0.0979
0.1488	9.0	27	0.0976
0.1375	10.0	30	0.0885
0.1375	11.0	33	0.0907
0.1375	12.0	36	0.0863
0.1375	13.0	39	0.0843
0.1297	14.0	42	0.0833
0.1297	15.0	45	0.0840
0.1297	16.0	48	0.0861
0.1241	17.0	51	0.0903
0.1241	18.0	54	0.0891
0.1241	19.0	57	0.0876
0.1185	20.0	60	0.0868

Framework versions

• Transformers 4.35.2
• Pytorch 2.2.1+cu121
• Datasets 2.18.0
• Tokenizers 0.15.2