INDEX

Explanations

minimizing civilian collateral damage

New Auto-Interp

Configuration

Prompts (Dashboard)

238,145 prompts, 512 tokens each

Dataset (Dashboard)

lmsys + oasst1

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

奇异

0.43

 Injection

0.42

 들어

0.41

 креп

0.41

 ব্যায়াম

0.40

engi

0.40

 गड

0.39

栏

0.39

 இள

0.39

 완성

0.39

POSITIVE LOGITS

 civilian

1.45

 civilians

1.38

 indiscriminate

1.18

 Civilian

1.16

 indiscrimin

1.09

 innocent

1.08

 innoc

0.89

 বেসামরিক

0.87

 humanitarian

0.87

 bystanders

0.82

Activations Density 0.011%