INDEX

Explanations

command line or terminal

The main thing this neuron does is detect references to using a command‐line interface (e.g. “command prompt,” “terminal,” “cmd”).

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

</sup>

-1.33

 clube

-1.27

 metody

-1.27

 konfigura

-1.27

StackNavigator

-1.27

 potem

-1.26

 amada

-1.26

sendKeys

-1.26

 knapp

-1.23

 zbyt

-1.23

POSITIVE LOGITS

 mandatario

1.24

☘

1.20

 aktiven

1.19

殇

1.16

 ツリー

1.14

 робота

1.13

but

1.12

iapkan

1.11

 until

1.10

 trium

1.10

Activations Density 0.189%