INDEX

Explanations

clause

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

 Clause

-1.67

 clause

-1.66

clause

-1.45

 clauses

-1.39

Clause

-1.39

 Clauses

-1.37

clauses

-0.97

 cláus

-0.95

CLA

-0.79

 Jefus

-0.66

POSITIVE LOGITS

OGND

0.55

AccessorTable

0.54

0.52

 onCancelled

0.50

 تضيفلها

0.46

 denn

0.46

en

0.46

ylo

0.45

le

0.45

et

0.44

Activations Density 0.025%