INDEX

Explanations

mathematical sequences, functions, and spaces

The neuron responds to high‐level mathematical nouns denoting abstract structures—e.g. “group,” “sequence,” or “map.”

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

crapers

-1.28

unno

-1.25

$-$

-1.25

是怎么回事

-1.24

carus

-1.24

celand

-1.24

の絵

-1.23

苼

-1.23

vermectin

-1.23

不是很

-1.18

POSITIVE LOGITS

in

1.82

of

1.68

 that

1.60

 udang

1.48

 which

1.44

 this

1.43

 whose

1.42

 takk

1.41

 اخیر

1.41

1.40

Activations Density 0.041%