INDEX

Explanations

phrases related to scientific research methodologies

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

bÃŃ

-0.08

buz

-0.07

isko

-0.07

Ngh

-0.07

piel

-0.07

Ø±ÙĬØ¹

-0.07

 assort

-0.07

eya

-0.06

tiler

-0.06

usz

-0.06

POSITIVE LOGITS

ix

0.07

 æĹ

0.07

 Ling

0.06

athan

0.06

 Moral

0.06

AUSE

0.06

 xuyÃªn

0.06

 novel

0.06

-test

0.06

iker

0.06

Activations Density 0.020%