INDEX

Explanations

occurrences of a specific probe or marker within the data

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

sizeCache

-0.81

\]

-0.56

IGENCE

-0.55

subpackage

-0.52

awtextra

-0.51

이터

-0.50

Orsay

-0.49

CAPTION

-0.49

;#

-0.49

__((

-0.48

POSITIVE LOGITS

probe

2.64

PROBE

1.63

Probe

1.35

 Probe

0.78

ftagPool

0.77

 كومونز

0.59

AutoScale

0.57

 بيها

0.57

MLLoader

0.56

bootstrapcdn

0.55

Activations Density 0.004%