INDEX

Explanations

mathematical variables and functions

This neuron activates on numeric literals—integers or decimal numbers appearing in formulas or code.

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

 itself

-1.77

 himself

-1.55

簠

-1.46

 своем

-1.39

 zustand

-1.33

own

-1.32

 BECAUSE

-1.30

vuo

-1.29

itself

-1.29

</h4>

-1.24

POSITIVE LOGITS

 Both

1.46

There

1.42

много

1.40

 Despite

1.30

 other

1.27

 ambos

1.23

 Unlike

1.20

 första

1.19

 Currently

1.17

 Sadly

1.16

Activations Density 0.050%