INDEX

Explanations

Big followed by a number

The neuron detects mentions of “Big” followed by a numeral (e.g. “Big Four,” “Big Six,” “Big Three”) used to label top groups or categories.

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

onOptions

-1.55

initializeApp

-1.55

 where

-1.52

kowania

-1.52

oceros

-1.42

,:);

-1.41

uteurs

-1.39

纡

-1.37

</u>

-1.34

."

-1.34

POSITIVE LOGITS

 whom

1.60

of

1.57

who

1.33

↓

1.26

without

1.25

Схема

1.25

のも

1.21

 типа

1.20

 detest

1.19

have

1.18

Activations Density 0.028%