INDEX

Explanations

honest admissions and opinions

The neuron activates on hedging or stance-taking phrases that introduce personal opinions or uncertainty (e.g. “I think,” “to be honest,” “we thought that,” “I’m not sure”).

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

ৢ

-1.55

ଃ

-1.48

 Khusus

-1.44

 claramente

-1.37

 Terbaru

-1.32

ﮦ

-1.31

 húmedo

-1.28

 hain

-1.25

langsung

-1.23

 Pastikan

-1.23

POSITIVE LOGITS

 only

1.56

 their

1.52

 many

1.35

 even

1.33

}=\

1.25

«

1.23

 tämän

1.19

 mesmo

1.19

Даже

1.18

».

1.16

Activations Density 0.045%