INDEX

Explanations

Canterbury dairy farm

The neuron activates on numeric literals in code, picking out sequences of digits (with or without decimal points).

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

 that

-2.08

or

-1.83

 this

-1.59

 usual

-1.52

側は

-1.51

 бампер

-1.49

></

-1.46

 και

-1.41

to

-1.38

 […]

-1.38

POSITIVE LOGITS

fordable

1.54

 ainfi

1.51

 erbjud

1.36

 QUIZ

1.32

 POOL

1.30

 MONTH

1.29

Politik

1.29

 Louvain

1.28

Gang

1.27

 wzgl

1.27

Activations Density 0.001%