INDEX

Explanations

pupil on

This dataset/training/program

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

 timings

0.55

 GetData

0.53

 ateliers

0.52

 pours

0.52

 anisotropy

0.51

 identifiers

0.51

 angrily

0.51

 eateries

0.50

 rekonstru

0.50

 classifiers

0.50

POSITIVE LOGITS

0.57

żeń

0.55

Ford

0.51

Fluor

0.51

Francis

0.50

Young

0.49

partnership

0.49

Brom

0.48

Partners

0.48

Whole

0.47

Activations Density 0.001%