INDEX

Explanations

Summer Heat, Quit Smoking, In Situ

The main thing this neuron does is spot standalone numeric tokens (in particular the digit “1”).

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

 most

-1.51

in

-1.48

 sehr

-1.43

 енер

-1.35

 when

-1.31

 whose

-1.30

it

-1.28

all

-1.27

 after

-1.27

 epitom

-1.22

POSITIVE LOGITS

 alguns

1.36

 Dinas

1.30

 !!!!

1.27

蹰

1.23

 jezt

1.21

 légèrement

1.18

 soooo

1.18

 remarkably

1.18

!!!!!!!

1.16

unn

1.16

Activations Density 0.000%