INDEX

Explanations

list items separated by comma

The neuron fires on standalone numeric literals (tokens consisting of digits) in code snippets.

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

is

-1.78

 bolsillos

-1.61

 that

-1.46

 cejas

-1.41

 länder

-1.39

you

-1.36

ḻ

-1.33

 originalmente

-1.32

 pestañas

-1.27

arschu

-1.27

POSITIVE LOGITS

 його

1.45

 наоборот

1.38

Its

1.34

to

1.30

 with

1.28

에도

1.27

 niektórych

1.26

 scelte

1.25

 飯店

1.24

 pôr

1.23

Activations Density 0.156%