INDEX

Explanations

that means

The neuron specifically detects occurrences of the word “that,” regardless of context or casing.

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

uttgart

-1.98

 мы

-1.96

atação

-1.94

was

-1.85

ୈ

-1.81

鯪

-1.78

ௗ

-1.77

cleros

-1.71

 настоящий

-1.70

蜑

-1.70

POSITIVE LOGITS

2.25

Its

2.14

“

2.02

1.83

the

1.81

Various

1.72

 скорее

1.70

 szív

1.69

КС

1.69

интере

1.69

Activations Density 0.041%