INDEX

Explanations

variations of the word "can" and its negations in different 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

RIX

-0.07

zcze

-0.07

gence

-0.07

olars

-0.07

hta

-0.06

loth

-0.06

 gratuita

-0.06

Ð²ÐµÑģÑĤÐ¸

-0.06

ATS

-0.06

APT

-0.06

POSITIVE LOGITS

Ø±Ø®

0.07

’t

0.07

_cpp

0.07

't

0.07

º

0.06

 Lump

0.06

rab

0.06

ÑĢÐ¾Ð²Ð¾Ð´

0.06

itational

0.06

haz

0.06

Activations Density 0.020%