INDEX

Explanations

use this file

The neuron primarily picks up occurrences of the word “use” (as in boilerplate license text “you may not use…”).

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

íferos

-0.72

மாக

-0.71

UFACT

-0.70

CONVERT

-0.70

cets

-0.69

טרה

-0.68

诣

-0.67

 Niveau

-0.67

vre

-0.66

imą

-0.66

POSITIVE LOGITS

EMG

0.75

rücke

0.67

illen

0.66

tamide

0.65

Emma

0.62

/*

0.62

 Nouveau

0.62

(:,:,

0.61

gzip

0.61

 danken

0.60

Activations Density 0.076%