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

    Neuronpedia

    INDEX
    Explanations

    is fundamentally

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     случайно
    0.73
    械
    0.71
     Fairchild
    0.65
     boredom
    0.62
    LessThan
    0.62
     frustrations
    0.61
    injlim
    0.60
    嘈
    0.60
     qualche
    0.60
    蒨
    0.59
    POSITIVE LOGITS
     deeply
    1.27
     ethically
    0.98
     fraught
    0.89
     profundamente
    0.85
     profondément
    0.84
     morally
    0.83
     antit
    0.82
    deep
    0.82
     tantamount
    0.81
     against
    0.80
    Act Density 0.229%

    No Known Activations