INDEX

Explanations

find greatest common divisor

This neuron detects prominent action verbs in code comments or documentation strings—especially the imperative verbs that start descriptions of what a function does.

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

憷

-1.77

apparently

-1.61

 kompet

-1.61

Garantie

-1.61

 atrak

-1.60

 vijf

-1.59

seems

-1.59

appears

-1.58

karakter

-1.57

 divertida

-1.55

POSITIVE LOGITS

 from

2.23

 with

1.95

for

1.83

by

1.82

of

1.78

to

1.74

at

1.68

on

1.63

or

1.62

 provided

1.62

Activations Density 0.049%