INDEX

Explanations

classifying items into types

The neuron activates on broad, abstract, uncountable nouns (e.g. “weather,” “nature,” “functions,” “spirituality,” “activity,” “energy,” “senses,” etc.).

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

may

0.73

0.68

 capacitance

0.64

 laptop

0.60

 mattress

0.59

 court

0.57

ability

0.57

 }^{\

0.57

0.56

 paternity

0.55

POSITIVE LOGITS

slaught

0.79

 เลย

0.77

 全部

0.74

 どれ

0.73

 종합

0.73

 แยก

0.72

 모두

0.72

 Utils

0.71

 раздел

0.71

 گیا۔

0.70

Activations Density 0.263%