INDEX

Explanations

defining parts using where

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

 since

0.41

 puisqu

0.40

 recalling

0.40

 imagines

0.40

 우리는

0.39

เพราะ

0.39

हमने

0.39

We

0.39

因為

0.39

 puisque

0.38

POSITIVE LOGITS

Breaking

0.75

 breaking

0.73

break

0.71

breaking

0.71

 break

0.68

 broken

0.67

 breakdown

0.66

broken

0.66

 Breaking

0.66

 breaks

0.64

Activations Density 0.022%