INDEX

Explanations

The neuron fires strongly on single-quote (apostrophe) characters.

possessive 's or year

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

Some

-1.38

 revisor

-1.32

♩

-1.23

他们

-1.22

Sometimes

-1.19

他们是

-1.19

 lamenta

-1.18

 reaksi

-1.17

有些人

-1.16

 abus

-1.15

POSITIVE LOGITS

咎

1.34

vian

1.31

irmanship

1.24

achio

1.23

 قوات

1.22

ophobic

1.21

sians

1.20

KommentareTeilen

1.19

Verlag

1.17

缉

1.17

Activations Density 0.003%