INDEX

Explanations

convention or prevention

The neuron detects references to formal international treaties or conventions.

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

ায়

-0.84

🛺

-0.82

 LEFT

-0.82

kac

-0.81

голь

-0.80

onix

-0.80

 varargin

-0.79

 głów

-0.77

已有

-0.77

 pertanian

-0.77

POSITIVE LOGITS

racist

0.80

Mig

0.79

 naše

0.78

 podpis

0.78

ھے

0.77

 destacar

0.77

Mig

0.77

 racism

0.77

 모든

0.76

学位

0.76

Activations Density 0.018%