INDEX

Explanations

code and markup delimiters

This neuron detects isolated, standalone tokens that serve as structural separators or delimiters (e.g. a lone “break;”, a lone HTML tag, a comment terminator, or a single-line quotation marker).

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

uoja

-1.16

 également

-1.15

 также

-1.13

 Также

-1.11

 También

-1.05

kuun

-1.05

ですが

-1.04

 oblasti

-1.03

izadores

-1.02

ۦ

-1.02

POSITIVE LOGITS

 will

0.91

Before

0.90

One

0.90

They

0.87

↵

0.86

 bądź

0.85

Each

0.85

↵↵

0.84

Keep

0.83

 such

0.82

Activations Density 0.000%