INDEX

Explanations

phrases related to various conflicts and wars

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

ÎµÏĨ

-0.07

ioned

-0.07

hone

-0.07

crest

-0.07

Bullet

-0.07

 Alive

-0.07

alli

-0.07

irty

-0.06

typings

-0.06

ingredient

-0.06

POSITIVE LOGITS

 terror

0.15

 Terror

0.13

 drugs

0.13

 Drugs

0.12

 terrorism

0.12

 Terrorism

0.11

terror

0.11

 Terra

0.10

 terra

0.10

terra

0.10

Activations Density 0.012%