INDEX

Explanations

prepositions indicating direction or recipient

The neuron strongly activates on words that signal clarification or instruction (e.g. clarify, expectations, obvious, clear).

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

 woodland

0.50

டுவது

0.50

谤

0.50

 нуля

0.50

 Salisbury

0.49

 деклара

0.49

SED

0.49

ებით

0.48

 bilmi

0.48

ocomplete

0.48

POSITIVE LOGITS

 kepada

2.59

แก่

1.89

ให้กับ

1.80

 στους

1.73

aos

1.67

 لهم

1.59

에게

1.49

 людям

1.46

들에게

1.41

 детям

1.40

Activations Density 0.798%