INDEX

Explanations

instances of the word "oops" and other expressions of mistakes or errors

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

hal

-0.07

illisecond

-0.07

ebb

-0.07

Cancelable

-0.07

ev

-0.06

evt

-0.06

logue

-0.06

td

-0.06

upa

-0.06

andas

-0.06

POSITIVE LOGITS

aken

0.08

AGR

0.07

esity

0.06

es

0.06

/errors

0.06

ingly

0.06

/no

0.06

 Fallen

0.06

apers

0.06

 Cush

0.06

Activations Density 0.002%