INDEX

Explanations

references to recent years in scholarly or technical contexts

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

########.

-0.69

DockStyle

-0.65

Kjelder

-0.64

 douleurs

-0.54

DeleteBehavior

-0.54

pentry

-0.54

AddWithValue

-0.53

atterns

-0.52

ofe

-0.52

obed

-0.51

POSITIVE LOGITS

::~

0.65

UnsafeEnabled

0.65

HomeController

0.62

 rospy

0.54

BASELINE

0.52

stateProvider

0.52

simpleType

0.51

 AllAfrica

0.50

 meanwhile

0.50

 Theſe

0.49

Activations Density 0.012%