INDEX

Explanations

I followed by verb

This neuron fires on first-person references—specifically the pronoun “I” (and its immediate context in “I am …”).

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

_{

-1.46

-1.45

เพราะ

-1.40

某些

-1.30

 algunos

-1.29

mathrm

-1.29

-1.28

其他

-1.27

เป็น

-1.27

hline

-1.26

POSITIVE LOGITS

of

1.50

it

1.46

 PHOTOGRAPHY

1.39

thargy

1.39

的同时

1.33

ถม

1.28

the

1.28

It

1.27

rvats

1.25

"?

1.20

Activations Density 0.420%