INDEX

Explanations

possessive pronouns followed by nouns

New Auto-Interp

Configuration

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

eve

-0.08

epoch

-0.08

 dodge

-0.07

entr

-0.07

me

-0.07

 furn

-0.07

 internals

-0.07

omer

-0.07

(),

-0.07

LogLevel

-0.06

POSITIVE LOGITS

nas

0.13

 ###↵

0.13

Own

0.13

cen

0.13

self

0.12

SELF

0.12

 Bare

0.11

rut

0.11

 Butt

0.11

 Burl

0.11

Activations Density 0.006%