INDEX

Explanations

both GPT and Your

This neuron flags personal references—especially first- and second-person pronouns and possessives (I, me, my, you, your, our)—addressing the speaker or reader.

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

ილ

0.58

ilo

0.53

 হোটেলে

0.49

했을

0.49

ilitas

0.48

emorrh

0.47

sex

0.47

 기능을

0.47

をお願い

0.46

 cere

0.46

POSITIVE LOGITS

 myself

2.39

 ourselves

2.36

 Myself

2.31

 yourself

2.22

 Yourself

2.19

 himself

2.17

自分

2.16

 oneself

2.13

 себе

2.02

me

2.00

Activations Density 0.475%