INDEX

Explanations

technical terms and media types

This neuron fires on numerical measurements and technical design specifications in the text.

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

or

-2.02

ଶ

-1.60

 will

-1.45

but

-1.38

 response

-1.32

éu

-1.28

民众

-1.25

ּוֹ

-1.23

Zunanje

-1.23

to

-1.20

POSITIVE LOGITS

ׇ

1.70

ﻻ

1.48

ﻷ

1.47

 nien

1.40

ις

1.36

ló

1.36

râ

1.35

 bouteille

1.34

 النبات

1.30

 flèche

1.30

Activations Density 0.011%