INDEX

Explanations

guide and guides

The main thing this neuron does is find occurrences of the word “guide” (including its plural “guides”).

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

-2.19

↵↵

-1.98

↵

-1.84

-1.63

-1.49

 да

-1.48

ième

-1.45

-1.44

 что

-1.42

人で

-1.40

POSITIVE LOGITS

geſ

1.86

 حين

1.74

狆

1.66

😰

1.59



1.55

тельность

1.54

EFAULT

1.53

 Haaland

1.52

ктором

1.51

tifications

1.48

Activations Density 0.005%