INDEX

Explanations

lists, design, or tasks

The neuron fires on the main action or descriptive phrase that heads each item in listicle‐style content (e.g. “Avoid,” “Establish a mission,” “Easy to clean,” “Feed your dog properly,” etc.).

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

 also

-1.25

喏

-1.23

:'#

-1.18

omiast

-1.09

って感じ

-1.06

 vías

-1.05

.”.

-1.03

<>

-1.02

 segredo

-1.02

ÍTICA

-1.01

POSITIVE LOGITS

 that

1.21

 your

1.20

şil

1.09

 салат

1.05

 أو

1.05

adakan

1.03

İK

1.03

Either

1.02

 zarówno

1.02

 veya

1.01

Activations Density 0.020%