INDEX

Explanations

describing state and appearance

This neuron detects numeric tokens (including integers and decimals).

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

for

-1.50

 with

-1.49

to

-1.46

 like

-1.43

WRENCE

-1.30

 chociaż

-1.26

-1.17

These

-1.16

 across

-1.16

NOT

-1.15

POSITIVE LOGITS

 itself

1.59

も行

1.45

its

1.45

قى

1.45

 vårt

1.42

 valgt

1.40

㎖

1.39

 abierta

1.38

;\\

1.38

 sellado

1.38

Activations Density 0.123%