INDEX

Explanations

law enforcement reports

The neuron fires on verbs and terms that describe criminal actions—especially theft and related offenses—in police‐report style prose.

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

our

-1.78

we

-1.49

 mengumumkan

-1.38

歟

-1.34

 putra

-1.24

 bakteri

-1.22

 ACKNOWLEDGMENTS

-1.14

 staten

-1.13

我们

-1.11

≙

-1.11

POSITIVE LOGITS

德國

1.46

 وكان

1.38

pedro

1.30

 mężczy

1.27

日閲覧

1.23

 felicitación

1.22

 described

1.21

 photos

1.20

ragal

1.17

 Polskiej

1.16

Activations Density 0.021%