INDEX

Explanations

kiss and related words

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

 houses

-0.10

Øµ

-0.10

aina

-0.09

LEAN

-0.09

 doubling

-0.09

ikip

-0.09

FAST

-0.09

Opening

-0.09

 Governments

-0.09

POSITIVE LOGITS

(es

0.18

 kiss

0.13

 goodbye

0.12

 kissing

0.12

 kissed

0.11

wort

0.11

 Kiss

0.11

 kisses

0.11

able

0.10

 forehead

0.10

Activations Density 0.005%