INDEX

Explanations

android acl components

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

if

-1.82

 nová

-1.80

for

-1.75

пят

-1.74

皛

-1.71

だという

-1.65

ТП

-1.63

 siguen

-1.58

驺

-1.57

 trajes

-1.55

POSITIVE LOGITS

<bos>

1.74

an

1.69

</i>

1.66

uk

1.64

 credi

1.60

ur

1.60

 nece

1.60

简约

1.55

to

1.54

ak

1.54

Activations Density 0.005%