INDEX

Explanations

evidence based research

New Auto-Interp

Configuration

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 الحره

-0.72

 للاسماء

-0.68

RegressionTest

-0.68

 iprot

-0.64

FunctionFlags

-0.62

 utafitiHapana

-0.60

ineno

-0.59

 redonda

-0.58

 arşivlendi

-0.58

 seguintes

-0.58

POSITIVE LOGITS

')['

0.54

 doch

0.52

AnchorStyles

0.51

 didReceive

0.51

}}"

0.50

"]').

0.49

songwriter

0.47

[@

0.47

bandcamp

0.45

)['

0.45

Activations Density 0.690%