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

    Neuronpedia

    INDEX
    Explanations

    who questions about identity

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    的是
    0.70
    checkpoint
    0.68
     Check
    0.66
    {'
    0.65
    idir
    0.64
     is
    0.63
    asion
    0.63
    findung
    0.63
    Check
    0.62
     Territory
    0.62
    POSITIVE LOGITS
     najbol
    0.90
     buckles
    0.85
     melhores
    0.83
     બનાવી
    0.82
     mejores
    0.82
    騾
    0.81
    ربعة
    0.80
     joyas
    0.79
    玩意
    0.79
    yrıca
    0.78
    Act Density 0.065%

    No Known Activations