DantheMan124
/

deberta-preference-reward

Text Classification

preference-learning

Model card Files Files and versions

deberta-preference-reward / config.json

Commit History

Add reward_train_max_length: 512 so the calibration regime ships with the weights

00a4b81
verified

DantheMan124 commited on Jul 28

Upload DebertaV2ForSequenceClassification

fb6e0f5
verified

DantheMan124 commited on Jul 10