INDEX

Explanations

actions or possibilities

The neuron detects first-person statements describing or imagining where the speaker lives (e.g. “I lived on 50 acres in Montana…”).

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

are

-1.22

vább

-1.20

 рекоменду

-1.18

ּוֹ

-1.16

报记者

-1.12

now

-1.09

 mosca

-1.08

 your

-1.06

ַּ

-1.05

estaba

-1.05

POSITIVE LOGITS

 would

1.67

 preferably

1.49

 kanskje

1.35

Would

1.33

 gjerne

1.27

preferably

1.26

 olabilir

1.22

 vielleicht

1.21

my

1.16

嘁

1.15

Activations Density 0.022%