INDEX

Explanations

he, she, his, her

The neuron is detecting third-person personal pronouns (e.g. he, she, his, her, their) in the text.

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

ation

-1.30

at

-1.23

だら

-1.22

hô

-1.21

↵↵

-1.18

NOT

-1.16

 nossos

-1.16

ac

-1.16

response

-1.13

zepte

-1.12

POSITIVE LOGITS

 three

1.49

 when

1.38

six

1.29

 seven

1.27

 sebelum

1.25

now

1.24

нига

1.24

月号

1.23

two

1.23

 Când

1.20

Activations Density 0.044%