INDEX

Explanations

Workers

New Auto-Interp

Configuration

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ValueStyle

-0.70

localctx

-0.52

MessageOf

-0.51

 transférez

-0.51

 ویکی‌پدی

-0.50

MergeFrom

-0.50

yeh

-0.49

 صوتيه

-0.49

tagHelper

-0.48

存于互联网档案馆

-0.47

POSITIVE LOGITS

 Workers

0.65

 workers

0.59

 Worker

0.56

 worker

0.56

urlencoded

0.56

Workers

0.55

Bef

0.54

 WORKERS

0.52

 surgeries

0.52

 drop

0.51

Activations Density 0.001%