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

    Neuronpedia

    INDEX
    Explanations

    various contexts

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     spectateurs
    -0.72
    tamiento
    -0.70
     craindre
    -0.69
     Simulated
    -0.67
    譯
    -0.66
     capacité
    -0.65
     kip
    -0.65
    ።
    -0.64
     rost
    -0.64
     Multivariate
    -0.64
    POSITIVE LOGITS
     requests
    0.76
     Spanien
    0.73
    」，
    0.71
     Statements
    0.69
     Head
    0.68
    calyptic
    0.68
     quên
    0.68
    وانید
    0.68
     Abteilung
    0.68
    high
    0.68
    Act Density 0.045%

    No Known Activations