INDEX

Explanations

any or anything

The neuron detects question‐forming auxiliary verbs and related words (e.g. “Did,” “Was,” “Have,” “Any”) that signal an interrogative sentence.

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

❔

-1.29

hhhhh

-1.20

 ??????

-1.15

???:

-1.15

asted

-1.14

ooooooo

-1.13

²/

-1.13

罍

-1.12

љено

-1.12

taya

-1.11

POSITIVE LOGITS

any

3.52

 anything

2.55

 herhangi

2.03

or

1.93

any

1.89

 anywhere

1.81

 ever

1.75

 quelconque

1.67

anything

1.64

Any

1.63

Activations Density 0.050%