INDEX

Explanations

motor and motivation

The neuron consistently lights up on tokens starting with the root “mot-” (e.g. motor, motivated, motive).

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

ᱨ

0.56

鞏

0.55

 विज्ञ

0.54

htmlspecialchars

0.54

Це

0.54

 आकर्षण

0.52

ائح

0.52

য়ান

0.52

 Zaragoza

0.52

 يتعلق

0.52

POSITIVE LOGITS

Mot

1.04

Mot

1.00

mot

0.97

mot

0.82

 motifs

0.80

MOT

0.77

 मोट

0.77

MOT

0.74

 motives

0.74

 motivate

0.73

Activations Density 0.121%