INDEX

Explanations

phrases that indicate a procedure or set of instructions

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

–↵↵

-0.07

AVC

-0.07

hod

-0.07

hood

-0.07

hud

-0.07

sonian

-0.07

kop

-0.07

asal

-0.07

.construct

-0.07

asaki

-0.07

POSITIVE LOGITS

 follows

0.08

cert

0.07

 below

0.06

 seguint

0.06

EndElement

0.06

 novel

0.06

 ëĭ¤ìĿĮê³¼

0.06

 MaterialApp

0.06

ëĵľë¦¬

0.06

å¾Ģ

0.05

Activations Density 0.007%