INDEX

Explanations

that followed by explanatory clauses

This neuron detects question cues—i.e. interrogative words and phrases (who, what, how, when, why, etc.) that signal a question.

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

$]$

-1.59

óricos

-1.30

lerim

-1.27

textura

-1.25

 ktorá

-1.20



-1.20

,),

-1.19

of

-1.17

Жит

-1.17

-1.15

POSITIVE LOGITS

 kvinder

1.21

canakan

1.15

omenclature

1.14

cestors

1.13

 силь

1.09

 אוי

1.07

 Вы

1.07

 silueta

1.07

 strength

1.06

 paradis

1.06

Activations Density 0.107%