INDEX

Explanations

common phrase starters

The neuron spots words written in all capital letters (i.e. fully uppercase tokens).

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

-1.24

 seemed

-1.23

myapplication

-1.21

Hå

-1.16

-1.11

—

-1.10

 there

-1.09

unjuk

-1.07

celular

-1.06

-1.05

POSITIVE LOGITS

Ï

1.31

StructEnd

1.30

Ë

1.23

!,

1.20

ㅣ

1.19

Ą

1.16

Ħ

1.16

 speci

1.16

_;

1.14

へ向

1.14

Activations Density 0.056%