INDEX

Explanations

asking questions

whether a sentence is explicitly labeling or discussing a question (especially tokens marking "question", "direct/embedded question", or an explicit question phrase).

New Auto-Interp

Configuration

Prompts (Dashboard)

238,145 prompts, 512 tokens each

Dataset (Dashboard)

lmsys + oasst1

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

MethodBeat

0.35

ーツ

0.34

 الحل

0.34

ታል

0.33

固

0.33

త్రిక

0.32

linejoin

0.32

 patches

0.31

 coating

0.31

estock

0.31

POSITIVE LOGITS

 질문

2.83

问

2.81

 questions

2.78

 spør

2.78

 hỏi

2.77

 pregunta

2.75

 pergunta

2.70

 asking

2.69

 question

2.63

 întreb

2.63

Activations Density 0.196%