INDEX

Explanations

auxiliary verbs

words related to being informed or instructed

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

Negative Logits

learned

-0.59

 Learned

-0.51

chtete

-0.50

 missed

-0.49

UnknownFields

-0.49

MemoryWarning

-0.48

'">

-0.47

ScopeManager

-0.47

譜

-0.46

 learned

-0.46

POSITIVE LOGITS

 told

1.29

 asked

1.03

 encouraged

1.02

 instructed

1.02

 given

1.01

 informed

1.00

 advised

0.99

 Told

0.99

told

0.98

 reminded

0.97

Activations Density 2.131%