INDEX

Explanations

frequently asked

The neuron is triggered by structured alphanumeric codes and delimiters (e.g. uppercase letters, digits, and hyphens) commonly used as field tags or identifiers in reference and metadata sections.

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

-1.31

-1.21

it

-1.20

 versi

-1.20

but

-1.17

су

-1.13

意思是

-1.13

me

-1.12

rierten

-1.11

-1.08

POSITIVE LOGITS

 kış

1.28

 प्रत्येक

1.21

 pikir

1.20

⚋

1.19

utsu

1.18

 firstly

1.18

 Feuchtigkeit

1.18

𝓽

1.18

 propia

1.17

تها

1.16

Activations Density 0.103%