INDEX

Explanations

conditions or requirements

The neuron detects “where” clauses that introduce contextual or conditional situations in documentation.

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

ּוֹ

-1.22

 escolher

-1.13

瞟

-1.13

ستاگرام

-1.06

ִּ

-1.05

 兔子

-1.03

ⅿ

-1.02

 varier

-1.01

があれば

-1.00



-1.00

POSITIVE LOGITS

 want

1.56

or

1.48

 cần

1.44

ต้องการ

1.40

需要

1.39

 need

1.38

 require

1.26

 such

1.22

 required

1.20

;%%

1.13

Activations Density 0.136%