INDEX

Explanations

Code operators and negative

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

1.57

−

1.52

−

1.47

-(

1.43

$-

1.40

-\

1.32

-(

1.27

$-\

1.27

$-

1.25

-$

1.23

POSITIVE LOGITS

}=(-

0.47

)=(-

0.44

}}=-\

0.40

 subpo

0.38

恶

0.37

 negatively

0.36

 hated

0.35

 doubted

0.35

 খারাপ

0.34

 horrible

0.34

Activations Density 0.004%