INDEX

Explanations

physical contact or impact

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

vic

-0.10

 surve

-0.10

 Thorn

-0.10

ippo

-0.10

 Rosen

-0.09

 Layers

-0.09

 Hollywood

-0.09

 shar

-0.09

 Cary

-0.09

 weapons

-0.09

POSITIVE LOGITS

 against

0.20

against

0.16

 Against

0.13

Against

0.13

 contre

0.12

 holes

0.10

 hole

0.10

imas

0.10

 marks

0.09

 patter

0.09

Activations Density 0.013%