INDEX

Explanations

class of year

The neuron detects mentions of an academic “class” or cohort (e.g. graduating class, freshman class, class of …).

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

к

-1.64

<strong>

-1.54

amb

-1.51

score

-1.48

-1.45

if

-1.43

initial

-1.42

റ്

-1.42

 doua

-1.40

dicos

-1.39

POSITIVE LOGITS

pantalon

1.77

of

1.71

mudah

1.61

lampu

1.59

 Schicks

1.54

nicole

1.52

bracelet

1.51

igens

1.50

pantal

1.47

moins

1.45

Activations Density 0.040%