INDEX

Explanations

mathematical expressions and symbols

This neuron activates on numeric tokens that appear as parts of arithmetic expressions or equations (e.g. numbers next to operators in math formulas).

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

 sensing

0.50

 fixation

0.50

 polarization

0.48

 quadrature

0.47

 inductively

0.47

 radians

0.46

 oscillation

0.46

 iteratively

0.46

 generates

0.45

 telemetry

0.45

POSITIVE LOGITS

 manhã

0.51

 Espanha

0.51

 Confira

0.51

 verlieren

0.50

 Gucci

0.49

 aliados

0.49

 inverno

0.49

 Sabrina

0.48

 Juventus

0.48

 Inglaterra

0.48

Activations Density 0.000%