INDEX

Explanations

phrases that indicate significant size, quality, or impact in various contexts

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

ippo

-0.09

ÙĪØ§Ø±

-0.08

orders

-0.06

ble

-0.06

ï½¢

-0.06

BLE

-0.06

çĽĺ

-0.06

Ð°ÑĢÐ°Ð¼

-0.06

 èŃ

-0.06

.BackgroundColor

-0.06

POSITIVE LOGITS

 manner

0.13

 nutshell

0.10

 fashion

0.08

 hurry

0.07

 folks

0.07

 setting

0.06

icans

0.06

 sposÃ³b

0.06

xad

0.06

 vain

0.06

Activations Density 0.077%