INDEX

Explanations

names

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

rette

-0.85

 Lancet

-0.83

Ges

-0.80

 Roskov

-0.59

offsetof

-0.57

 Derek

-0.57

ipan

-0.54

industan

-0.54

_))

-0.53

ienced

-0.53

POSITIVE LOGITS

 Martyr

0.56

 Nauk

0.54

Jules

0.54

 Martyrs

0.53

Jú

0.52

 pleaſure

0.52

 OFDb

0.51

 Bartholomew

0.50

 gend

0.50

 Cune

0.49

Activations Density 1.430%