INDEX

Explanations

dates and case numbers

This neuron primarily detects the hyphen character as used within numerical identifiers (e.g. docket or case numbers).

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

は

-1.27

–

-1.27

-1.20

と

-1.18

を

-1.13

や

-1.12

がある

-1.03

が

-1.02

も

-1.01

 Wiel

-1.00

POSITIVE LOGITS

為に

1.27

久し

1.17

ところに

1.12

 esque

1.07

だいぶ

1.06

נין

1.05

割と

1.04

itarianism

1.03

 vorgenommen

1.02

Voltaje

1.01

Activations Density 0.120%