INDEX

Explanations

instances of negative feedback or criticism

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

place

-0.09

acific

-0.07

endon

-0.07

ISMATCH

-0.07

æł·çļĦ

-0.07

 é¦Ļ

-0.07

Ø³ÙħØ©

-0.07

iente

-0.07

rer

-0.07

onn

-0.07

POSITIVE LOGITS

/null

0.09

-negative

0.08

_INFINITY

0.08

/n

0.07

rones

0.07

IntegerField

0.07

-positive

0.06

ities

0.06

/problem

0.06

_integer

0.06

Activations Density 0.012%