INDEX

Explanations

of the

This neuron detects legal code citations or statutory reference phrases (e.g. “of this section/title”).

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

 것은

-1.16

one

-1.10

quilo

-1.09

all

-1.07

il

-1.05

他又

-1.04

some

-1.02

 concess

-1.02

 residuals

-1.01

 veritable

-1.01

POSITIVE LOGITS

the

1.30

РО

1.16

 черен

1.14

 expliqué

1.11

geç

1.10

 japoneses

1.09

ᐠ

1.09

ДО

1.08

bánh

1.06

 اندروید

1.06

Activations Density 0.013%