INDEX

Explanations

clearly distinguishable

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

OOT

-0.10

 bouquet

-0.10

 Pure

-0.09

 Demo

-0.09

amt

-0.09

 bart

-0.09

bart

-0.09

 simple

-0.09

 honeymoon

-0.09

 overst

-0.09

POSITIVE LOGITS

 clearly

0.20

 definitely

0.17

 identifiable

0.16

 recognizable

0.16

 Clearly

0.15

 unmistak

0.15

 explicitly

0.14

 recogn

0.14

Clearly

0.14

 distinctly

0.14

Activations Density 0.719%