INDEX

Explanations

provide input text

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 không

1.08

 возможно

1.08

 nhưng

0.98

Und

0.97

 perhaps

0.96

 हालांकि

0.94

but

0.93

but

0.93

 maybe

0.93

 না

0.93

POSITIVE LOGITS

實際

1.22

 actuality

1.15

 actual

1.11

 eigent

0.99

 structurally

0.99

 ACTUAL

0.98

 internals

0.98

实际

0.96

 STRU

0.96

 resten

0.95

Activations Density 1.237%