INDEX

Explanations

specifically, particularly, both

This neuron fires on discourse markers and parenthetical or enumerative cues—words and phrases like “i.e.,” “e.g.,” “both,” “consisting of,” “in particular,” and similar signals that introduce examples, clarifications, or transitions.

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

two

-1.73

its

-1.59

one

-1.57

 both

-1.52

-1.46

Ṭ

-1.45

 three

-1.40

on

-1.39

 only

-1.37

on

-1.30

POSITIVE LOGITS

 которые

1.73

 Mereka

1.66

 których

1.65

 которых

1.61

 themselves

1.60

 onları

1.49

 exceptionnelle

1.49

 véritables

1.49

 élevée

1.46

 różne

1.46

Activations Density 0.314%