INDEX

Explanations

abstract concepts followed by relation

The neuron activates strongly on content-bearing nouns and adjectives (key “informational” words) rather than on function 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

 will

-1.43

;}

-1.30

ждая

-1.22

engertian

-1.18

ных

-1.15

 HttpNotFound

-1.11

ūsų

-1.07

to

-1.05

;"></

-1.05

ferenz

-1.05

POSITIVE LOGITS

 интерьере

1.18

Hogyan

1.16

 книге

1.15

رفته

1.14

 basit

1.13

 deteriorating

1.11

 WITH

1.09

ويات

1.07

 сообщает

1.06

頂きました

1.06

Activations Density 0.166%