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

    Neuronpedia

    INDEX
    Explanations

    scripting video and livestream content

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     beeindruck
    0.49
    ذج
    0.47
    ع
    0.44
     roche
    0.42
    ו
    0.42
     belated
    0.41
     box
    0.41
    να
    0.41
     الدفاع
    0.40
    ટ
    0.40
    POSITIVE LOGITS
    cycles
    0.44
    fr
    0.42
    frist
    0.42
    ଫ
    0.42
    Executed
    0.41
    updates
    0.41
    fab
    0.40
    Storing
    0.40
     Updates
    0.40
    $-,
    0.40
    Act Density 0.001%

    No Known Activations