INDEX

Explanations

the adverb "as" used in various contexts

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

egra

-0.07

Rap

-0.06

owo

-0.06

avra

-0.06

oples

-0.06

udent

-0.06

trys

-0.06

VERY

-0.05

åľ³

-0.05

 Toolkit

-0.05

POSITIVE LOGITS

close

0.11

 close

0.11

 much

0.09

-close

0.09

warm

0.08

phy

0.08

lox

0.07

 likely

0.07

Close

0.07

olutely

0.07

Activations Density 0.017%