INDEX

Explanations

code

New Auto-Interp

Configuration

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

WriteBarrier

-0.64

TagHelpers

-0.60

<bos>

-0.56

 propOrder

-0.53

Carriera

-0.51

RegressionTest

-0.50

 excru

-0.48

Suara

-0.48

enterOuterAlt

-0.47

PreferredItem

-0.47

POSITIVE LOGITS

 mapStateToProps

0.71

InjectAttribute

0.59

LookAnd

0.57

 humaine

0.56

 ioutil

0.53

[](

0.53

новниш

0.52

atmosfera

0.50

 satélite

0.49

 themselves

0.48

Activations Density 0.178%