INDEX

Explanations

specific numerical or statistical references within the context of optimization processes

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

ulan

-0.07

abc

-0.07

á»ĳ

-0.06

 incumbent

-0.06

 Lite

-0.06

eler

-0.06

rossover

-0.06

ounge

-0.06

Lite

-0.06

 ÑģÑĥÑĤ

-0.06

POSITIVE LOGITS

 newer

0.09

 recent

0.08

 newly

0.08

 recently

0.08

annel

0.07

 exceptions

0.07

exceptions

0.07

 Rare

0.07

Rare

0.06

 seldom

0.06

Activations Density 0.032%