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

    Neuronpedia

    INDEX
    Explanations

    programming constructs and design patterns

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     ionic
    0.76
     cultivating
    0.76
     cultivate
    0.74
     cultivation
    0.73
     bleed
    0.73
     November
    0.73
     wavy
    0.73
     Great
    0.71
     endanger
    0.71
     ½
    0.71
    POSITIVE LOGITS
    Factory
    1.54
    Builder
    1.51
    Wrapper
    1.49
    Manager
    1.47
    Helper
    1.45
    Base
    1.35
    Controller
    1.34
    类
    1.33
    Info
    1.26
    Utils
    1.26
    Act Density 0.440%

    No Known Activations