INDEX

Explanations

the word "active" (or perhaps "actively")

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 washable

-0.69

 dica

-0.68

Vea

-0.68

 PartialEq

-0.66

wash

-0.63

ValueChanged

-0.63

áh

-0.63

activation

-0.62

HIR

-0.62

lassification

-0.61

POSITIVE LOGITS

 actively

1.13

 active

0.98

ImageContext

0.83

actively

0.82

 Active

0.81

 ACTIVE

0.80

Active

0.79

 presence

0.78

active

0.76

 ativo

0.76

Activations Density 0.015%