INDEX

Explanations

code/technical language

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

enic

-0.93

abit

-0.91

'"

-0.66

 Yard

-0.56

 Houſe

-0.55

’”

-0.54

OPE

-0.53

 Makan

-0.52

<?>>

-0.52

 behalf

-0.51

POSITIVE LOGITS

AndEndTag

0.71

 AssemblyProduct

0.69

 Roskov

0.65

 ModelExpression

0.64

featureID

0.60

CrossRef

0.59

IsContent

0.57

TypedDataSet

0.57

setVerticalGroup

0.54

PYX

0.52

Activations Density 1.634%