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

    Neuronpedia

    INDEX
    Explanations

    that defied, clawed, could, felt

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     fuit
    0.83
    '
    0.83
    य्
    0.79
    的表现
    0.78
    和一个
    0.78
    是个
    0.77
    ’
    0.76
    是一种
    0.75
    是一個
    0.75
    werben
    0.74
    POSITIVE LOGITS
     heretofore
    1.08
     otherwise
    0.98
     hitherto
    0.94
     Until
    0.93
    至今
    0.93
     stubbornly
    0.92
     Otherwise
    0.90
     until
    0.90
     simultaneously
    0.89
    迄
    0.88
    Act Density 0.021%

    No Known Activations