INDEX

Explanations

2

New Auto-Interp

Configuration

Prompts (Dashboard)

16,384 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

piac

-0.52

AddTagHelper

-0.50

IsContent

-0.49

 maternelle

-0.49

 šķ

-0.48

 wireType

-0.47

Hochspringen

-0.47

matchCondition

-0.47

③

-0.46

 Italij

-0.46

POSITIVE LOGITS

featureID

0.61

Zah

0.57

 Akbar

0.52

digest

0.49

 Usman

0.49

Kel

0.48

 Saba

0.48

 Shiva

0.48

hibian

0.48

Mab

0.47

Activations Density 0.069%