INDEX

Explanations

titles and age

This neuron detects the uppercase surnames (last names) that head obituary or death‐notice entries.

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

 però

-1.07

 similar

-1.04

 particular

-1.02

istical

-0.97

べ

-0.96

 itself

-0.95

 світу

-0.95

 powiedzieć

-0.93

fatti

-0.89

Các

-0.88

POSITIVE LOGITS

age

2.91

of

2.47

Age

1.71

Age

1.63

 ages

1.56

 âge

1.52

 edad

1.41

age

1.37

AGE

1.35

AGE

1.34

Activations Density 0.030%