INDEX

Explanations

phrases followed by verbs or actions

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

-0.14

’

-0.12

 contend

-0.12

lez

-0.11

Ih

-0.11

had

-0.10

.Selenium

-0.10

umbn

-0.10

IFn

-0.10

PED

-0.10

POSITIVE LOGITS

wan

0.10

 herself

0.09

 Barcl

0.09

me

0.09

 hers

0.09

 Serv

0.08

 himself

0.08

oc

0.08

 ener

0.08

 Gilles

0.08

Activations Density 0.017%