INDEX

Explanations

slightly, pronounced, profound, rapid, common

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

 equally

-0.11

 both

-0.11

 BOTH

-0.11

 ALSO

-0.10

 Hoch

-0.09

 inclusive

-0.09

 also

-0.09

azzi

-0.09

 unlike

-0.09

enge

-0.09

POSITIVE LOGITS

 slightly

0.12

 associates

0.11

lad

0.10

trad

0.09

dn

0.09

 poster

0.09

iam

0.09

former

0.09

shell

0.09

 Former

0.09

Activations Density 0.013%