INDEX

Explanations

anecdotes and stories

This neuron responds to prose that reads like a book‐or article‐style narrative or historical synopsis, especially when it names an author, title, or key dates.

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

 கூட்டணிய

0.63

概念

0.56

 nouve

0.51

concepts

0.50

☰

0.50

 ಸೂಚ

0.48

 conceito

0.48

verters

0.47

ารา

0.47

뾰

0.47

POSITIVE LOGITS

 memoirs

2.08

 memoir

1.98

 recounting

1.97

 anecdotes

1.96

 stories

1.95

 autobi

1.95

 autobiography

1.92

 relatos

1.89

 recounts

1.88

 racconto

1.81

Activations Density 0.468%