INDEX

Explanations

newest + noun

The neuron fires on mentions of something being the “newest” or latest release/version.

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

 afuera

-2.38

``

-2.31

-2.23

),

-2.20

 maxim

-2.19

ቍ

-2.05

 defunct

-2.02

 extremadamente

-1.99

 nuestros

-1.97

 pinnacle

-1.96

POSITIVE LOGITS

of

2.58

 inserir

2.23

 メンテナンス

2.23

 可愛い

2.19

 ayakkab

2.16

 kaybet

2.14

＝＝＝＝

2.11

2.09

exels

2.08

وين

2.08

Activations Density 0.004%