INDEX

Explanations

The neuron spotlights quoted object‐property names (i.e. keys) in JSON/JavaScript‐style data structures.

object or array properties

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

 this

-1.59

for

-1.54

 their

-1.40

cartier

-1.38

 from

-1.37

or

-1.35

gamis

-1.30

you

-1.28

 even

-1.27

 first

-1.24

POSITIVE LOGITS

Themen

1.38

一系列

1.34

举动

1.33

 aceste

1.33

 aproximativ

1.30

냑

1.29

 ejecutar

1.27

 sugerencias

1.26

 erwähnten

1.24

anniversaire

1.23

Activations Density 0.001%