INDEX

Explanations

code punctuation and symbols

The neuron activates on numeric string literals used as keys or identifiers (e.g. digits in quoted object/dictionary keys).

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

 will

-1.20

 ideen

-1.15

视着

-1.14

の方も

-1.13

 sidan

-1.04

providedIn

-1.04

 @"";

-1.01

えっ

-0.98

 each

-0.96

ёг

-0.96

POSITIVE LOGITS

 hugely

1.09

ेंस

0.99

 Cæsar

0.98

 overly

0.97

 decidedly

0.96

質感

0.96

 wacky

0.96

など

0.95

書籍

0.95

 quirky

0.94

Activations Density 0.020%