INDEX

Explanations

rade

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

 mode

-1.16

mode

-0.89

 Mode

-0.84

 modes

-0.73

 MODE

-0.73

Mode

-0.62

MODE

-0.60

encodeWith

-0.60

QMetaType

-0.58

denn

-0.57

POSITIVE LOGITS

 Houſe

0.82

 itſelf

0.78

 Jefus

0.71

 Chriſt

0.69

 ་་

0.69

 Monfieur

0.69

 houſe

0.68

 stiefel

0.68

 ſmall

0.68

 Shakspeare

0.68

Activations Density 0.944%