К содержимому
onsuggest
Войти
Статья en

Fun toy result: a subliminal-learning effect via on-policy distillation

Источник: LessWrong ·

In the standard subliminal learning setup, a teacher model with a preference generates data according to its policy, and then the student is trained on those data. I wondered if it was necessary for the teacher model to actually get to control the produced tokens, so I instead let a no-prompt "teacher" model produce the sequences to be trained on. Then I trained the student against a subliminal-learning-prompted teacher model's logits on those sequences. (Since an unmodified "teacher" is identical to the student model, this is equivalent to just doing on-policy distillation from the prompted t

Что такое Fun toy result: a subliminal-learning effect via on-policy distillation?

Fun toy result: a subliminal-learning effect via on-policy distillation is a статья discovery curated from LessWrong. In the standard subliminal learning setup, a teacher model with a preference generates data according to its policy, and then the student is trained on those data. I wondered if it

Как получить доступ или попробовать?

You can experience this content directly on its original host (LessWrong) by clicking the "Open Original" button.

Откуда этот контент?

Curated from LessWrong by the OnSuggest team to surface the most interesting and serendipitous content on the web.

Горячие клавиши

Предыдущее открытие
←P
Следующее открытие
→N
Нравится
L
Не для меня
X
Сохранить
S
Показать эту справку
?
Закрыть
Esc

Горячие клавиши работают на странице открытий.