INDEX

Explanations

punctuation marks, particularly commas and quotation marks

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

ditor

-0.07

 zwar

-0.07

burg

-0.06

cola

-0.06

ucas

-0.06

ypes

-0.06

hurst

-0.06

pref

-0.06

rint

-0.06

nerg

-0.06

POSITIVE LOGITS

 \↵

0.07

Ð»ÐµÐ¶

0.06

":""

0.06

ophe

0.06

à¤Ĺà¤²

0.06

vais

0.06

forcer

0.06

ÐµÑı

0.06

.hasOwnProperty

0.05

tcb

0.05

Activations Density 0.011%