INDEX

Explanations

quantified durations in days

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

膊

-1.96

？）

-1.85

 doivent

-1.82

"));

-1.78

 работают

-1.77

{}'.

-1.77

 rivales

-1.73

嫪

-1.65

 spéciaux

-1.65

癀

-1.63

POSITIVE LOGITS

Väl

1.82

我们

1.74

你

1.65

1.64

—

1.59

比率

1.57

习近平

1.57

 superbly

1.57

 unparalleled

1.57

 fearing

1.57

Activations Density 0.009%