INDEX

Explanations

The neuron detects alumni degree credentials—abbreviations like BA, BS, MBA—followed by an apostrophe and graduation year.

graduation years

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

it

-1.32

 manifesta

-1.23

他想

-1.09

他现在

-1.08

 после

-1.07

 opie

-1.05

}}}

-0.99

 bens

-0.95

orElse

-0.95

its

-0.95

POSITIVE LOGITS

 traditionnels

1.12

Sly

1.05

 akár

1.03

や

1.02

 voidaan

1.01

 różnych

0.99

who

0.98

0.97

 pourquoi

0.95

 eccellente

0.94

Activations Density 0.002%