INDEX

Explanations

scientific 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

ç»Łè®¡å±Ģ

-0.27

UNS

-0.26

è´¸

-0.26

éĴ´

-0.25

 competitions

-0.24

ummer

-0.24

 STATS

-0.23

.beginPath

-0.23

yes

-0.23

åĺ

-0.23

POSITIVE LOGITS

ankan

0.31

RICT

0.28

çıŃç»Ħ

0.28

ç®¡çĲĨå±Ģ

0.27

ÑģÑĤÐ²ÐµÐ½Ð½ÑĭÐµ

0.26

uggest

0.25

.setColumns

0.25

agedList

0.25

 threaded

0.24

ãģĬãģĭ

0.24

Activations Density 0.882%