INDEX

Explanations

words followed by uncertain outcomes

This neuron responds to subjective or modal expressions—words that convey frequency (often, rarely), desire or opinion (would like, enjoy, think), and similar hedging language.

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

and

-2.95

 сказала

-1.66

 ((((

-1.43

 fourteen

-1.38

((*

-1.34

stwa

-1.34

 four

-1.33

 with

-1.32

 וגם

-1.31

 oraz

-1.30

POSITIVE LOGITS

 talvez

1.59

its

1.56

 tentar

1.55

 poate

1.45

 fizeram

1.41

说不定

1.41

 hizo

1.34

 mesmo

1.34

就没

1.34

菒

1.31

Activations Density 0.133%