INDEX

Explanations

specific purposes or qualities

The neuron activates on numeric expressions (especially multi-digit numbers and decimals) in the text.

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

or

-1.66

 являются

-1.16

ländischen

-1.09

on

-1.09

䊐

-1.09

as

-1.08

 السل

-1.05

🎣

-1.05

 یک

-1.03

 gelb

-1.03

POSITIVE LOGITS

 більше

1.66

 určit

1.47

ctrons

1.36

 ruchu

1.34

avons

1.33

ureux

1.31

[])

1.30

 však

1.29

 nemá

1.29

 nhưng

1.27

Activations Density 0.072%