INDEX

Explanations

colors and associated nouns

The neuron identifies color words (e.g. “red,” “yellow,” “blue”) in the text.

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

 bicarbonate

-0.86

 определить

-0.83

ượt

-0.82

isesti

-0.82

 golden

-0.79

 colourful

-0.78

 uterine

-0.77

湲

-0.77

teira

-0.77

 zvíř

-0.76

POSITIVE LOGITS

 haired

1.25

='\

0.97

washing

0.92

bird

0.92

 اللون

0.91

 tinge

0.88

hirts

0.87

prints

0.86

wash

0.85

 tint

0.84

Activations Density 0.066%