INDEX

Explanations

phrases that suggest careful consideration or caution

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

rena

-0.06

sip

-0.06

amin

-0.06

elt

-0.06

ORIZ

-0.06

edn

-0.05

AGO

-0.05

ATORY

-0.05

ð

-0.05

el

-0.05

POSITIVE LOGITS

 fact

0.12

 factor

0.10

Factor

0.09

 dikkate

0.09

 Factor

0.09

 Fact

0.08

.factor

0.08

 ÑĥÑĩÐ¸ÑĤ

0.08

 consideration

0.08

fact

0.08

Activations Density 0.023%