INDEX

Explanations

no or negative statements

The neuron fires on hedging and limiter expressions—light modal verbs (could, have, need), minimal quantifiers (little, no), and phrases like “as yet,” “no further,” or “know of” that soften or qualify a statement.

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

 jajaja

-1.75

in

-1.74

🎶

-1.66

of

-1.63

´)

-1.63

at

-1.61

~~~~~

-1.60

²)

-1.59

[])

-1.59

 oiseau

-1.58

POSITIVE LOGITS

 neither

2.11

no

2.08

 nothing

2.05

mathrm

2.02

don

2.00

 keine

1.89

not

1.84

 ничего

1.83

 none

1.83

 keinen

1.80

Activations Density 0.155%