INDEX

Explanations

Kota Kinabalu

The neuron detects long hexadecimal identifier strings (e.g., GUIDs or hex‐encoded addresses).

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

 though

-0.82

 never

-0.81

 schon

-0.76

 blick

-0.75

never

-0.74

Never

-0.73

તી

-0.73

 Never

-0.73

 exakt

-0.72

planned

-0.71

POSITIVE LOGITS

تحميل

0.75

 according

0.75

传递

0.74

Aal

0.73

PES

0.73

 Tanto

0.72

 zählen

0.71

ODI

0.71

ľov

0.71

Ave

0.69

Activations Density 0.109%