INDEX

Explanations

legal or policy terms including officials and groups

The neuron is primarily detecting numeric and statistical expressions (numbers, percentages, decimal values, etc.).

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

 modulo

0.75

'));

0.66

 Unisex

0.66

 velcro

0.65

 euclidean

0.65

 gestalt

0.62

 datatype

0.62

 Euclidean

0.60

 একখানি

0.58

POSITIVE LOGITS

 lawmakers

0.95

 prosecutors

0.85

 dissidents

0.85

 mismanagement

0.82

 policymakers

0.80

 militants

0.78

 تهد

0.78

 protesters

0.77

 legislators

0.77

 traffickers

0.77

Activations Density 0.037%