INDEX

Explanations

personal pronouns (you, me, us)

The neuron strongly activates on first‐ and second‐person personal pronouns (e.g. “I,” “me,” “you”), i.e. direct self‐ or reader‐address.

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

rassed

-1.53

鲥

-1.46

＇

-1.41

澌

-1.39

菇

-1.36



-1.35

 expédié

-1.34

 enfance

-1.34

 באמצעות

-1.33

 vnnd

-1.31

POSITIVE LOGITS

as

1.45

at

1.42

1.41

may

1.38

to

1.37

 having

1.34

 suggests

1.34

1.33

and

1.32

 from

1.30

Activations Density 0.033%