INDEX

Explanations

technical terms and specific metrics relating to system performance and memory management

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

blr

-0.07

.scalablytyped

-0.07

 somehow

-0.06

Ø·Ùģ

-0.06

kop

-0.06

å°ĩ

-0.06

kili

-0.06

.Lang

-0.06

Lim

-0.06

oter

-0.06

POSITIVE LOGITS

 unless

0.19

unless

0.17

 depending

0.15

 Unless

0.15

Unless

0.15

 typically

0.13

depending

0.13

 usually

0.13

 Depending

0.11

Depending

0.11

Activations Density 0.055%