INDEX

Explanations

pseudo or so-called claims

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 trustworthy

-1.27

 honest

-1.15

 properly

-1.14

 accidentally

-1.14

 important

-1.13

 reputable

-1.12

 reliable

-1.09

 decently

-1.09

 correct

-1.09

 legendary

-1.07

POSITIVE LOGITS

所谓

1.70

 prétend

1.52

 pseudo

1.40

pseudo

1.35

voer

1.25

 supuestamente

1.14

所谓的

1.14

 centaines

1.13

まるで

1.13

phony

1.12

Activations Density 0.208%