INDEX

Explanations

tokens after `The` or `//`

The neuron fires on any token containing numeric digits (i.e. numbers or alphanumeric strings with digits).

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

in

-1.47

at

-1.29

by

-1.27

 brought

-1.25

one

-1.24

 quite

-1.23

 facilitating

-1.23

 just

-1.21

all

-1.21

as

-1.21

POSITIVE LOGITS

mainly

1.40

primarily

1.39

particularly

1.39

athered

1.38

usually

1.28

increase

1.27

 soooo

1.27

 différents

1.26

 verschillende

1.25

mostly

1.22

Activations Density 0.001%