INDEX

Explanations

pronouns: I, you, we

The neuron fires on personal address pronouns—especially “you,” “I,” and “we.”

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

-1.63

遽

-1.45

ль

-1.31

maxx

-1.30

 persones

-1.29

»,

-1.25

xvii

-1.23

 âmes

-1.23

брав

-1.20

ник

-1.19

POSITIVE LOGITS

もう一度

1.86

没有什么

1.60

1.48

有两

1.45

色々

1.42

あっという

1.39

もう少し

1.36

没能

1.35

 Konzentration

1.35

much

1.35

Activations Density 0.037%