INDEX

Explanations

do you have

This neuron detects occurrences of the verb “have” (and related “have any/you have” constructions) in question contexts, signaling user requests or offers of information.

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

不管是

-1.71

}$

-1.70

焼肉

-1.63

}{

-1.61

and

-1.56

 только

-1.52

多くの

-1.52

 fournis

-1.52

 myriad

-1.51

 multitudes

-1.48

POSITIVE LOGITS

’

1.80

 طويلة

1.75

in

1.69

 centavos

1.66

ꝛ

1.63

 Krzyż

1.52

 jeho

1.52

 garota

1.51

 Strö

1.50

</strong>

1.49

Activations Density 0.023%