INDEX

Explanations

Code and phrases

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

ValueStyle

-0.66

expandindo

-0.63

RectangleBorder

-0.50

QMetaType

-0.47

 saith

-0.46

ValueGenerated

-0.46

checkNotNull

-0.45

 Tower

-0.44

 Kobayashi

-0.44

WriteTagHelper

-0.44

POSITIVE LOGITS

 CreateTagHelper

0.60

στό

0.60

 ſtate

0.56

 pleaſure

0.56

@[+][

0.54

ammans

0.54

adpleegd

0.53

 raiſ

0.52

>[]

0.52

 purpoſe

0.52

Activations Density 0.001%