INDEX

Explanations

proper nouns and places

The neuron fires on section headings or subheadings in the document (e.g. “Early career,” “Steam Era,” “By Air,” etc.).

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

of

-2.23

-1.27

who

-1.25

咐

-1.18

 whose

-1.17

pageInfo

-1.17

 如果

-1.15

 paździer

-1.09

createdBy

-1.09

šo

-1.09

POSITIVE LOGITS

 there

1.35

揾

1.34

There

1.31

 trovo

1.31

ups

1.27

ykjav

1.26

ciare

1.26

 dolorosa

1.26

美国的

1.25

食べ物

1.23

Activations Density 0.174%