INDEX

Explanations

roles and professions

The neuron strongly activates on words naming people’s occupations or professional roles (e.g. “author,” “broadcaster,” “singer,” “entrepreneur”).

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

是一款

0.70

是一家

0.67

繁忙

0.65

的公司

0.64

三年

0.61

是在

0.59

)});

0.59

各类

0.59

是以

0.57

খিল

0.57

POSITIVE LOGITS

 lawmaker

1.10

 biographer

1.05

 columnist

1.03

 thinker

1.02

 statesman

1.02

 politician

1.01

 cricketer

1.01

 broadcaster

1.00

 commentator

0.98

 critic

0.96

Activations Density 0.300%