INDEX

Explanations

license or warranty disclaimers

This neuron fires on the minus sign in negative numeric expressions (e.g. the leading “–” in subscripts or limits indicating a negative number).

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

ஂ

-2.58

-2.56

 have

-2.38

at

-2.36

-2.25

is

-2.16

en

-2.16

-2.13

 went

-2.08

POSITIVE LOGITS

AutoresizingMask

2.64

 españoles

2.52

،

2.42

 opulent

2.41

的表情

2.38

ᨕ

2.36

 secteurs

2.31

 revered

2.31

 postpon

2.30

 Impressive

2.30

Activations Density 0.004%