INDEX

Explanations

We have

This neuron fires on instances of first-person plural pronouns (e.g. “we,” “our”) in the text.

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

 being

-1.25

 actually

-1.22

has

-1.20

 통해

-1.20

并且

-1.19

된

-1.18

 probably

-1.17

 also

-1.16

時間は

-1.16

Книга

-1.15

POSITIVE LOGITS

{*

1.45

 ſon

1.44

 ajouter

1.30

蓠

1.27

歺

1.27

 correspondant

1.26

you

1.23

)”

1.23

 perſon

1.22

 вас

1.21

Activations Density 0.122%