INDEX

Explanations

rearrange

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

arte

-0.59

 arte

-0.50

 esquí

-0.49

otel

-0.49

light

-0.47

 jorden

-0.47

quot

-0.45

 rearranged

-0.45

chkin

-0.45

riwal

-0.43

POSITIVE LOGITS

 Houſe

0.76

 Anſ

0.71

 imageNamed

0.69

setViewName

0.69

 Conſ

0.68

 houſe

0.68

 Diſ

0.68

AccessorTable

0.66

 Theſe

0.66

 transfieras

0.65

Activations Density 1.644%