INDEX

Explanations

references to evaluating situations or issues within specific contexts

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

avra

-0.06

ULER

-0.06

raf

-0.06

anje

-0.06

jde

-0.06

lej

-0.06

shint

-0.06

 Sherman

-0.06

 fizz

-0.05

uno

-0.05

POSITIVE LOGITS

 terms

0.09

terms

0.08

ossier

0.07

Ð½Ñı

0.07

çİ

0.07

ulerAngles

0.07

lace

0.07

.inst

0.06

ampoo

0.06

Activations Density 0.026%