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

    Neuronpedia

    INDEX
    Explanations

    moments of reflection or realization

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ن
    2.63
    ع
    1.95
    ס
    1.89
    h
    1.87
    a
    1.86
    an
    1.80
     notion
    1.80
    п
    1.64
    ר
    1.64
     mite
    1.63
    POSITIVE LOGITS
    パクト
    2.09
    extrémité
    2.08
    dır
    2.00
     itibaren
    1.97
    lardan
    1.94
    彭
    1.93
    arına
    1.92
    ına
    1.87
    یکه
    1.84
     samano
    1.84
    Act Density 0.013%

    No Known Activations