INDEX

Explanations

animal body parts

This neuron detects animal names and references, particularly cats, lions, snakes, and dogs.

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

iles

-0.73

 stocks

-0.71

 committed

-0.70

ometry

-0.70

 auft

-0.69

коля

-0.68

 étoit

-0.67

の出

-0.67

ILES

-0.67

花

-0.67

POSITIVE LOGITS

skin

1.19

foot

1.16

 foot

1.06

tail

1.00

skins

0.87

SKIN

0.87

head

0.87

Foot

0.86

FOOT

0.85

 tail

0.84

Activations Density 0.045%