INDEX

Explanations

assess

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

 assess

-1.23

assess

-1.16

assessment

-1.09

 assessing

-1.05

Assessment

-1.04

 assesses

-1.02

Assessing

-0.98

 Assess

-0.98

 Assessing

-0.98

Assess

-0.93

POSITIVE LOGITS

ThroughAttribute

0.66

 Theſe

0.57

 tightness

0.55

er

0.52

VIO

0.50

ício

0.50

Aig

0.50

Aos

0.49

orap

0.48

ctable

0.48

Activations Density 0.737%