INDEX

Explanations

conditional statements

This neuron responds to semantically-rich content words (mainly nouns and verbs) rather than to common function words.

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

ങ്കിലും

0.62

 để

0.61

 inom

0.60

었고

0.57

 rangka

0.57

 niin

0.56

했고

0.55

 ochr

0.55

 upang

0.54

 malay

0.54

POSITIVE LOGITS

 when

1.91

when

1.87

 quando

1.78

if

1.75

 जब

1.75

 wenn

1.74

gdy

1.70

 cuando

1.70

 когда

1.68

 عندما

1.68

Activations Density 0.904%