INDEX

Explanations

`$` followed by variable names

This neuron specializes in detecting the “@” sigil used in code snippets (e.g. Razor/HTML helpers or other at-sign–prefixed tokens).

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

着一个

-1.70

if

-1.55

an

-1.52

by

-1.51

着一

-1.47

最重要的

-1.44

 gehören

-1.43

-1.41

珖

-1.37

 می‌تواند

-1.34

POSITIVE LOGITS

’

1.77

 Soyez

1.62

 Tentang

1.60

 Août

1.59

 teraz

1.56

 Terbaru

1.52

 rabat

1.51

 Commencez

1.48

 latter

1.47

Activations Density 0.019%