INDEX

Explanations

go or goes questions

The neuron is triggered by three‐word sequences of the form “X goes Y,” such as “regex goes here” or “piece goes where.”

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

्ष

-0.81

asString

-0.78

 Gelegenheit

-0.77

 otwor

-0.77

ricted

-0.76

nią

-0.74

 availability

-0.73

ണ്ട

-0.73

signol

-0.73

ங்கள

-0.72

POSITIVE LOGITS

 goes

3.80

 Goes

2.75

go

2.73

goes

2.70

Goes

2.58

 went

2.27

 GOES

2.16

 belongs

1.95

 going

1.89

 belong

1.88

Activations Density 0.082%