INDEX

Explanations

buying or reviewing products

This neuron detects first-person testimonial language—especially the pronoun “I” (and related personal experience statements).

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

 бампер

-1.55

 przód

-1.50

 السياس

-1.38

 schmeckt

-1.38

ₒ

-1.29

}}}

-1.28

 الحكوم

-1.27

🏟

-1.24

动静

-1.21

 برخی

-1.20

POSITIVE LOGITS

 Amazon

1.44

 purchase

1.30

مزید

1.24

 purchased

1.23

 seller

1.16

Amazon

1.10

 product

1.09

 sini

1.08

THIS

1.08

 amazon

1.08

Activations Density 0.016%