INDEX

Explanations

non-standard characters and formatting typically found in technical texts or documents

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ãĥ³ãĥĩãĤ£

-0.07

unga

-0.07

Yug

-0.07

addle

-0.06

 Kodi

-0.06

ubat

-0.06

 onStop

-0.06

lep

-0.06

-"

-0.06

—"

-0.06

POSITIVE LOGITS

 Cambridge

0.10

 Rout

0.10

 Oxford

0.09

0.08

 isbn

0.08

 Printed

0.08

 paperback

0.08

 ISBN

0.08

 Editorial

0.08

 Princeton

0.08

Activations Density 0.004%