INDEX

Explanations

introducing examples or elaborations

The neuron flags function words and following tokens that introduce elaboration or examples—e.g. “such as …,” “including …,” “through …”

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

 flesta

-1.58

 these

-1.45

 calentamiento

-1.41

 senaste

-1.38

ꪻ

-1.36

 fleste

-1.25

 bakgrund

-1.20

ịch

-1.19

"":

-1.18

'':

-1.16

POSITIVE LOGITS

by

2.22

 through

1.50

via

1.46

 โดย

1.38

 include

1.30

 mediante

1.29

to

1.25

 with

1.25

use

1.23

 whether

1.20

Activations Density 0.061%