INDEX

Explanations

while or when

This neuron detects sentence-initial temporal or contrastive conjunctions (e.g. “While,” “Whilst,” “When”).

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

 Fabulous

0.64

 partecip

0.63

 Democr

0.61

mousemove

0.57

 Gradually

0.57

之时

0.56

brace

0.55

ស្ថានភាព

0.55

 escribir

0.55

DEFINE

0.55

POSITIVE LOGITS

searchText

0.57

linkColor

0.53

ंता

0.51

ScriptObject

0.50

性と

0.50

ステム

0.50

 बीसीजी

0.49

onError

0.49

 ты

0.49

 zebrafish

0.49

Activations Density 0.000%