INDEX

Explanations

serious harm or offensive impact

The neuron responds to words signaling severe harm or danger—terms denoting violence, serious injury, death, or threat to life.

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

 frau

-1.45

 organisme

-1.45

 prenn

-1.45

 signaler

-1.36

 herbes

-1.35

 pengar

-1.35

 kram

-1.34

 Profes

-1.33

 tirs

-1.30

 bahay

-1.30

POSITIVE LOGITS

or

3.98

 или

1.90

 either

1.84

 alebo

1.62

 hoặc

1.56

 либо

1.48

any

1.40

 other

1.33

 substantial

1.33

 people

1.31

Activations Density 0.017%