INDEX

Explanations

introducing consequences or transitions

This neuron detects formal discourse–marking adverbs or transitional words (e.g. hence, thus, ultimately, particularly) that signal emphasis or logical connection.

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

-1.63

-1.55

他にも

-1.49

ほか

-1.37

もあります

-1.35

↵↵

-1.34

-1.32

 stället

-1.30

 masing

-1.29

 использование

-1.28

POSITIVE LOGITS

2.02

 from

1.85

 your

1.72

 those

1.60

 that

1.57

 acogedor

1.49

the

1.48

who

1.47

 their

1.45

員

1.45

Activations Density 0.078%