INDEX

Explanations

commmas and quotes

The neuron is basically a “number detector” – it lights up on numeric literals (e.g. integers or floats) 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

 Secondo

-1.30

 Anche

-1.27

和

-1.27

ciutto

-1.18

entrySet

-1.13

 Ogni

-1.13

 like

-1.13

今日は

-1.11

了他的

-1.10

 detik

-1.09

POSITIVE LOGITS

1.41

ti

1.27

 tons

1.23

 tratan

1.20

pēc

1.19

 desarrollando

1.17

最后的

1.16

!')

1.15

’)

1.13

 erstre

1.11

Activations Density 0.124%