INDEX

Explanations

namely, in other words

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

1.65

Referensi

1.47

izing

1.42

ized

1.39

ired

1.34

************/

1.27

izer

1.26

ifes

1.26

orney

1.25

ised

1.24

POSITIVE LOGITS

즉

1.49

这就是

1.39

 핵심

1.37

 voila

1.35

 这是

1.34

 это

1.33

 μαγγ

1.31

 ultimate

1.30

 이것

1.29

 тобто

1.28

Activations Density 0.867%