INDEX

Explanations

The neuron activates on four-digit numerals representing model years (e.g. car manufacturing years).

car makes and models

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

 verschillende

-1.71

 diversos

-1.62

 diferentes

-1.55

 différentes

-1.52

 algunos

-1.51

 ส์

-1.50

擽

-1.46

 própri

-1.45

 marav

-1.41

 différents

-1.40

POSITIVE LOGITS

of

2.61

all

1.55

in

1.48

 because

1.42

on

1.41

 with

1.34

 using

1.33

 papir

1.32

 entirely

1.29

 exquisite

1.27

Activations Density 0.015%