INDEX

Explanations

switch case statements

The neuron detects numeric literals (tokens consisting of digits) 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

 burbu

-1.45

of

-1.37

amkeit

-1.27

 lørdag

-1.27

 сумка

-1.26

 fotográfico

-1.25

 регулярно

-1.24

 xadrez

-1.24

 criatura

-1.23

⿺

-1.23

POSITIVE LOGITS

1.26

 zwięks

1.18

on

1.14

1.09

↵

1.08

 подчер

1.07

 pathetic

1.06

見る

1.05

1.04

Activations Density 0.001%