INDEX

Explanations

ordinal numbers followed by nouns

The neuron fires on ordinal number words (e.g. “second,” “third,” “fourth,” “ninth,” 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

 mehrerer

-1.37

 alcune

-1.23

 diversos

-1.17

って言

-1.17

रोध

-1.16

CÓ

-1.16

새

-1.13

 alcuni

-1.13

素敵

-1.12

 unceasing

-1.12

POSITIVE LOGITS

if

1.47

)।

1.27

at

1.24

 than

1.20

</h1>

1.16

 aniversario

1.14

 when

1.11

 Octubre

1.10

˘

1.09

͞

1.09

Activations Density 0.020%