INDEX

Explanations

always ready

The neuron is a strong detector of rare or domain‐specific tokens—especially numbers, dates, proper nouns, abbreviations, and other out-of-vocabulary items.

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

ib

0.85

 deviate

0.78

 copied

0.78

ERE

0.77

వారి

0.76

 करी

0.76

ON

0.74

MN

0.74

RUN

0.74

TAB

0.73

POSITIVE LOGITS

نت

0.87

ަށް

0.79

卞

0.76

ramiento

0.71

 McManus

0.71

dió

0.70

必须

0.69

ໍ່

0.69

covers

0.69

javase

0.69

Activations Density 0.000%