INDEX

Explanations

articles starting with "we"

This neuron activates strongly on the document’s title or heading line, marking the paper’s title.

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

-1.98

on

-1.86

two

-1.73

and

-1.57

for

-1.56

-1.52

On

-1.41

-1.40

POSITIVE LOGITS

躄

1.91

 capot

1.73

 mág

1.72

 grotte

1.71

鬯

1.70

sekten

1.67

 fraude

1.65

 alarme

1.64

 lampe

1.63

 alpes

1.61

Activations Density 0.332%