INDEX

Explanations

so, followed by a word

The neuron detects the discourse marker “So,” (often at the start of a sentence) signaling a follow-up question or topic shift.

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

↵

-1.56

 descenso

-1.45

 ถ้า

-1.45

 posiada

-1.41

 molti

-1.37

 zowel

-1.36

 naszych

-1.36

 både

-1.34

 wielu

-1.33

ron

-1.31

POSITIVE LOGITS

ilets

1.56

new

1.52

 while

1.42

شاء

1.40

……」

1.37

 although

1.34

öl

1.33

 unrelenting

1.33

 despite

1.32

Activations Density 0.019%