INDEX

Explanations

mentions of Reddit and its related terms or interactions

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

åĢĻ

-0.07

erken

-0.06

holds

-0.06

 Elaine

-0.06

 deep

-0.06

inea

-0.06

kom

-0.06

 freeze

-0.05

POSITIVE LOGITS

abbo

0.08

ÑħÐ¾Ð²

0.08

ivate

0.07

zyst

0.07

/OR

0.07

coil

0.07

tember

0.07

OutOfRangeException

0.07

à¸ģ

0.06

.om

0.06

Activations Density 0.001%