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

    Neuronpedia

    INDEX
    Explanations

    algorithmic complexity and performance

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     figurines
    0.43
    userManager
    0.40
    規範
    0.40
    衞
    0.39
     consentimiento
    0.39
    ইসলাম
    0.39
     sociedades
    0.39
     কমন
    0.39
    竣工
    0.38
    鸷
    0.38
    POSITIVE LOGITS
     inefficient
    1.01
     worst
    0.95
     inefficiencies
    0.89
     inefficiency
    0.89
    Worst
    0.86
    worst
    0.86
     Worst
    0.86
     computationally
    0.84
     asymptotically
    0.83
     efficiency
    0.82
    Act Density 0.075%

    No Known Activations