INDEX

Explanations

knowing or not knowing

This neuron responds to words and morphemes expressing knowledge or ignorance (e.g. know, ignorant, let me know).

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

 Dép

-0.82

 GMAT

-0.75

ISSA

-0.74

 arXiv

-0.73

HasKey

-0.70

Η

-0.68

zol

-0.68

 Universidade

-0.68

 Grims

-0.68

UCI

-0.67

POSITIVE LOGITS

 know

1.80

 knew

1.43

 Know

1.41

알

1.39

 knows

1.36

KNOW

1.28

know

1.24

 KNOW

1.24

 knowledge

1.23

知

1.23

Activations Density 0.034%