Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    INDEX
    Explanations

    programming library components

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     [
    0.54
    0.52
     OK
    0.51
     noise
    0.50
     November
    0.49
     the
    0.49
     filename
    0.48
     Figure
    0.48
     Primary
    0.48
     (
    0.48
    POSITIVE LOGITS
    Builder
    0.78
    Factory
    0.73
    Manager
    0.72
    Adapter
    0.72
    Utils
    0.69
    Options
    0.69
    Helper
    0.67
    Wrapper
    0.64
    Accessor
    0.64
    Converter
    0.61
    Act Density 0.177%

    No Known Activations