INDEX

Explanations

promptly comply with immediately

questions and answers

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

antiene

0.55

ampaikan

0.53

avorable

0.50

<unused4>

0.49

<unused41>

0.49

 специалист

0.49

 районов

0.48

 พูด

0.46

 മത്സര

0.46

 वगैरह

0.46

POSITIVE LOGITS

 indiscriminate

0.51

 necrotic

0.47

 degeneration

0.46

 planks

0.45

 ingestion

0.45

 hull

0.44

 gale

0.44

 destroyed

0.43

 ingested

0.43

 indiscrimin

0.43

Activations Density 0.002%