INDEX

Explanations

tap actions and inputs

The neuron detects occurrences of the word “tap” (and its variant “taps”).

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

—

-3.13

↵↵

-2.69

 vollständige

-2.59

-2.50

 neue

-2.42

-2.38

‘

-2.36

‘

-2.30

 และ

-2.30

 größere

-2.27

POSITIVE LOGITS

ቺ

2.97

geſ

2.83

utiva

2.56

 sœur

2.55

 поверхности

2.48

ِّف

2.42

Ᏹ

2.41

鋨

2.39

 είναι

2.38

 próximas

2.34

Activations Density 0.010%