INDEX

Explanations

Color Profile

This neuron detects numeric expressions and related measurement or unit tokens (e.g. multi‐digit numbers, “6,” “4‐bit,” “monotonically increasing,” resolution figures, etc.).

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

 বিদ্রোহী

0.61

 التن

0.59

 সৃ

0.57

 আহ্ব

0.56

ிற்க

0.56

ிற்கு

0.55

 khen

0.55

 distra

0.54

 සො

0.54

劊

0.54

POSITIVE LOGITS

 uint

1.54

uint

1.49

 UInt

1.43

 unsigned

1.38

UInt

1.34

unsigned

1.33

UINT

1.20

 UINT

1.19

 uintptr

1.18

Uint

1.14

Activations Density 0.062%