INDEX

Explanations

from/in

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

 EconPapers

-0.84

 ProtoMessage

-0.84

 protoimpl

-0.80

 للمعارف

-0.80

NameInMap

-0.77

ReusableCell

-0.76

 ModelExpression

-0.75

 članak

-0.73

ьаж

-0.71

 Normdatei

-0.70

POSITIVE LOGITS

 power

0.58

 charge

0.56

power

0.50

 charged

0.47

Power

0.47

 Power

0.47

 bands

0.46

ek

0.45

 they

0.44

 their

0.44

Activations Density 0.030%