INDEX

Explanations

This neuron fires on first-person personal pronouns and related self-referential speech (e.g., “I,” “me,” “we”)—typically marking direct quoted speech or personal statements.

was it

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

طرز

-1.23

Currently

-1.23

 Mereka

-1.22

bedien

-1.21

szak

-1.20

gehe

-1.17

 terdengar

-1.13

denk

-1.12

舵

-1.09

Already

-1.08

POSITIVE LOGITS

was

1.92

 didn

1.33

 felt

1.32

 knew

1.28

 wasn

1.20

 wanted

1.19

 était

1.17

had

1.15

 gave

1.14

did

1.13

Activations Density 0.016%