INDEX

Explanations

confirmation seeking

The neuron detects confirmation‐seeking questions, especially short “Is that…?” or tag‐question constructions (e.g. “Is that correct?”, “Fair?”, “Right?”).

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

-1.33

🏢

-1.32

 recours

-1.28

 Япо

-1.28

 används

-1.23

ครื่อง

-1.23

 délicieux

-1.23

 berani

-1.21

 Buscar

-1.18

The

-1.17

POSITIVE LOGITS

1.77

 then

1.69

 this

1.52

伎

1.42

--

1.40

ابراین

1.33

 gereja

1.33

 ചെയ

1.32

 beck

1.27

 этого

1.25

Activations Density 0.039%