INDEX

Explanations

code declarations or definitions

This neuron detects numeric literals (numbers) in source code.

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

if

-1.66

as

-1.37

 while

-1.34

 because

-1.33

 when

-1.17

of

-1.10

 their

-1.07

 since

-1.04

-1.03

ятся

-1.00

POSITIVE LOGITS

Udo

1.01

 quantités

0.95

 bazı

0.95

 stalwart

0.95

 hermosos

0.91

Ofer

0.91

куртка

0.90

ߔ

0.90

めっちゃ

0.90

 maravill

0.89

Activations Density 0.009%