INDEX

Explanations

The main thing this neuron does is detect mentions of water.

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

les

0.68

stellungen

0.67

ട്ടുള്ള

0.66

崎

0.64

izons

0.64

 Flicker

0.63

িয়াছেন

0.63

利用者

0.63

UserProfile

0.63

expérience

0.62

POSITIVE LOGITS

 waters

1.48

 water

1.47

 воды

1.42

 wastewater

1.40

 slurry

1.40

 effluents

1.40

 wody

1.35

 água

1.34

 vody

1.32

 effluent

1.29

Activations Density 0.899%