INDEX

Explanations

crimes and related legal actions

The neuron detects mentions of crime-related terms (e.g. crimes, hate crimes, war crimes, sex offender).

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

-2.09

It

-2.05

 donned

-1.91

ſelves

-1.70

盘点

-1.64

The

-1.56

UNGEN

-1.53

IBRATION

-1.48

 anges

-1.46

 Франция

-1.46

POSITIVE LOGITS

ing

2.06

 zweiten

1.96

1.88

1.84

but

1.76

”,

1.73

”，

1.73

.”

1.71

蛩

1.69

逐渐

1.65

Activations Density 0.004%