INDEX

Explanations

variadic arguments

The neuron is triggered by code‐style punctuation and operator tokens (e.g. “*”, “,”, “**”, “(”, “)”, “...”) rather than regular words.

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

蟊

-1.80

 while

-1.69

if

-1.58

 YOUNG

-1.34

by

-1.34



-1.31

 novo

-1.30

 SIDE

-1.27

 aislada

-1.25

Ⅱ

-1.23

POSITIVE LOGITS

 Когда

1.35

或是

1.33

 nabízí

1.31

 patří

1.30

 wygląda

1.28

Когда

1.26

Кто

1.25

Kedves

1.24

as

1.21

首次

1.21

Activations Density 0.037%