INDEX

Explanations

new findings and concepts

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

—

-1.43

 ВИ

-1.31

tać

-1.30

);

-1.29

}`,

-1.28

 wartet

-1.27

$-$

-1.26

レード

-1.25

 announce

-1.24

 visión

-1.24

POSITIVE LOGITS

嗬

1.59

 häufiger

1.43

んだけど

1.41

 öfter

1.40

ですが

1.40

 esterni

1.39

 vermute

1.38

 länger

1.38

 This

1.36

<bos>

1.35

Activations Density 0.003%