INDEX

Explanations

kinetic energy

The neuron strongly activates on personal names—especially surnames (notably those beginning with “K”).

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

 вывести

0.57

Calls

0.57

ॅक

0.56

देवा

0.56

 malades

0.55

 Calls

0.55

ysteine

0.53

晛

0.52

答应

0.52

ولا

0.51

POSITIVE LOGITS

Ki

0.68

Ki

0.68

 Contents

0.60

KIT

0.59

KI

0.56

KP

0.55

яти

0.55

Kis

0.54

anç

0.54

Kue

0.54

Activations Density 0.270%