INDEX

Explanations

a relation or property

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

-2.08

ка

-1.91

le

-1.88

не

-1.73

<bos>

-1.71

より

-1.71

 что

-1.66

 verheiratet

-1.66

你觉得

-1.66

lized

-1.66

POSITIVE LOGITS

’

2.53

–

2.13

 When

2.05

--

1.86

if

1.82

 when

1.69

 neceffary

1.68

);

1.63

 kiedy

1.62

for

1.60

Activations Density 0.057%