INDEX

Explanations

label followed by colon

keyword followed by colon explanation

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

 \%$.

0.55

ಲೆಂ

0.50

ించాలని

0.49

കളും

0.48

ர்களையும்

0.48

ങ്ങളും

0.47

<unused432>

0.46

 বৎসর

0.45

രുവനന്തപു

0.45

<unused338>

0.44

POSITIVE LOGITS

1.66

：

1.55

1.28

ः

1.11

::

1.10

："

1.06

:...

1.05

：“

1.02

፡

1.00

*:

0.99

Activations Density 0.156%