INDEX

Explanations

making the world a better place

The neuron fires on key words that evoke broad social impact or improvement—terms about making the world, communities, health, lives, etc. better.

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

-1.42

to

-1.35

 unless

-1.29

but

-1.24

-1.13

 except

-1.07

But

-1.02

Ehrungen

-1.00

 ancak

-1.00

 gdyby

-0.98

POSITIVE LOGITS

 through

2.06

one

2.03

 poprzez

1.59

by

1.57

 while

1.55

 através

1.55

 melalui

1.44

 attraverso

1.34

 mediante

1.30

 wherever

1.28

Activations Density 0.079%