INDEX

Explanations

relevance

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

✨:

-0.71



-0.64

 "..\..\

-0.60

Vidite

-0.58

 CreateTagHelper

-0.57

 Vinc

-0.56

inSlope

-0.56

semantics

-0.55

 "..\..\..\

-0.54

eafter

-0.54

POSITIVE LOGITS

to

0.69

for

0.64

and

0.55

ValueStyle

0.53

rawValue

0.51

┻

0.51

 argint

0.50

DTD

0.49

 diminuer

0.49

 reducir

0.48

Activations Density 0.001%