INDEX

Explanations

murder and violent death

New Auto-Interp

Configuration

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ulan

-0.09

 escape

-0.09

 cons

-0.08

 Hannity

-0.08

irq

-0.08

 Damage

-0.08

 Fault

-0.08

HD

-0.08

 outlaw

-0.08

storm

-0.07

POSITIVE LOGITS

 murdered

0.37

 murder

0.29

 murders

0.28

 brutally

0.25

 mutil

0.25

 killed

0.25

 slain

0.24

 Murder

0.24

 brutal

0.21

 slaughtered

0.21

Activations Density 0.357%