INDEX

Explanations

The neuron fires on numeric constants (digits) in code or documentation.

Network penetration, solid phase, multi band

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

 deportivas

-1.23

 概要

-1.17

 fighting

-1.10

 ایس

-1.10

jąc

-1.08

椎

-1.07

ąc

-1.06

 :");

-1.05

 подоб

-1.05

İM

-1.04

POSITIVE LOGITS

or

1.60

Alerter

1.31

for

1.24

 насеље

1.20

 terbatas

1.20

 RBIs

1.18

komt

1.16

 estekak

1.13

pueden

1.13

use

1.12

Activations Density 0.010%