INDEX

Explanations

terms related to challenges and requirements for successful performance

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

/GPL

-0.07

formats

-0.07

enos

-0.07

Ease

-0.07

ehr

-0.07

ibur

-0.07

ardu

-0.06

unca

-0.06

uil

-0.06

æĪĺæĸĹ

-0.06

POSITIVE LOGITS

 skills

0.10

 skill

0.10

 strategy

0.08

 Skills

0.08

 strateg

0.08

skill

0.08

 Strategy

0.07

 strategies

0.07

skills

0.07

 skilled

0.07

Activations Density 0.033%