INDEX

Explanations

getter methods

It detects unusually long runs of leading whitespace (i.e. lines with deep indentation).

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

 rire

-1.42

 Traducción

-1.40

 individuelles

-1.34

が多く

-1.30

 organisme

-1.27

 boden

-1.27

 episódios

-1.26

 סס

-1.25

 investir

-1.23

 maig

-1.23

POSITIVE LOGITS

↵↵

1.77

</h2>

1.63

 what

1.52

</h1>

1.39

)=>{

1.38

==""){

1.38

 like

1.34

 from

1.27

by

1.27

get

1.27

Activations Density 0.002%