INDEX

Explanations

phrases or statements that emphasize the importance of rephrasing or clarifying concepts

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

hm

-0.08

AtPath

-0.07

fly

-0.07

voir

-0.07

Mgr

-0.07

iqu

-0.07

FileStream

-0.06

Indeed

-0.06

ãģ¾ãģŁ

-0.06

vas

-0.06

POSITIVE LOGITS

0.10

mith

0.08

0.07

sterol

0.07

ness

0.06

éĶĭ

0.06

onth

0.06

onec

0.06

AK

0.06

 kÄ±s

0.06

Activations Density 0.006%