INDEX

Explanations

complimentary, complacency, compliment

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

oved

-0.08

 prive

-0.07

 Tick

-0.07

 firma

-0.07

bed

-0.07

 Targets

-0.07

ÙĦØ§

-0.07

 Photos

-0.07

els

-0.06

 sized

-0.06

POSITIVE LOGITS

song

0.12

edb

0.10

lesh

0.10

bourne

0.10

fbe

0.10

edList

0.09

×Ļ

0.09

 Dere

0.09

 Leigh

0.09

fdf

0.09

Activations Density 0.001%

No Known Activations