INDEX

Explanations

prepositions and conjunctions

This neuron detects prepositions or adverbial phrases that introduce a basis or means (e.g. “through,” “based on,” “according to”).

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

 they

-2.16

They

-1.70

 вони

-1.47

THEY

-1.45

蛱

-1.45

 они

-1.42

 olímp

-1.41

 perché

-1.38

 Mereka

-1.34

емость

-1.32

POSITIVE LOGITS

 then

1.65

by

1.39

 that

1.38

—

1.34

者に

1.30

 with

1.21

 everybody

1.20

on

1.19

𝗕

1.16

 each

1.14

Activations Density 0.015%