INDEX

Explanations

form

New Auto-Interp

Configuration

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

çĴł

-0.24

owler

-0.24

 lawy

-0.24

 nÃ¤mlich

-0.24

 hitter

-0.23

 STDMETHODCALLTYPE

-0.21

.assertThat

-0.21

ftime

-0.20

å¦ĤæĦı

-0.20

æĸĩå¨±

-0.20

POSITIVE LOGITS

 form

2.00

å½¢å¼ı

1.70

form

1.45

çļĦå½¢å¼ı

1.43

 Form

1.30

_form

1.30

	form

1.24

Form

1.23

 forms

1.22

=form

1.18

Activations Density 1.053%