INDEX

Explanations

okay, as a transition

The neuron activates on the discourse marker “okay” (in any capitalization) whenever it appears 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

 całą

-2.34

 sieben

-2.33

-2.27

 usual

-2.13



-2.11

 heutige

-2.02

歛

-2.02

聍

-1.98

 setzt

-1.98

落ち着いた

-1.98

POSITIVE LOGITS

or

2.25

as

2.23

 uſed

2.06

𖣘

2.00

 手绘

1.99

 raiſ

1.96

-¡

1.95

媖

1.94

 BIBLIO

1.93

 faſt

1.90

Activations Density 0.004%