INDEX

Explanations

referring to prior information

all of the above

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

𝕞

0.39

াস্ত

0.37

ंख्य

0.36

டுத்த

0.33

𝕤

0.33

conder

0.32

যোগী

0.31

𝔾

0.31

defineProperty

0.31

<unused672>

0.31

POSITIVE LOGITS

 above

3.59

上述

3.44

above

3.33

 aforementioned

3.33

 выше

3.25

 aforesaid

3.13

上記

3.13

上記の

3.11

 mentioned

3.08

 acima

3.03

Activations Density 0.314%