INDEX

Explanations

section introductions

The neuron fires on numeric values (especially decimals) in the text—e.g. statistics, scores, or measurements.

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

 with

-1.63

for

-1.38

 While

-1.30

-1.24

 ensure

-1.23

 there

-1.23

 möchten

-1.23

on

-1.23

as

-1.21

̻

-1.21

POSITIVE LOGITS

 "..\..\

1.60

pography

1.49

澌

1.46

følge

1.45

%"),

1.45

錾

1.45

fand

1.43

🧐

1.41

XIII

1.41

☺☺

1.39

Activations Density 0.052%