INDEX

Explanations

names after and

The neuron fires strongly on the word “and” when it’s used to join items in a list (especially in enumerations of names).

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

szak

-1.36

👎

-1.34

 kompati

-1.25

 kontinu

-1.25

 respek

-1.24

 laku

-1.24

:{}

-1.22

moderne

-1.22

grupp

-1.21

......”

-1.17

POSITIVE LOGITS

 other

1.46

1.35

 others

1.34

1.20

1.17

1.15

1.13

 trusty

1.12

or

1.09

Activations Density 0.000%