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

    Neuronpedia

    INDEX
    Explanations

    code assignment or manipulation

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    -
    0.73
    ellä
    0.52
    ־
    0.52
     haberse
    0.51
    Está
    0.50
     Coca
    0.49
     crème
    0.47
     booze
    0.47
    ­
    0.47
    ებისა
    0.47
    POSITIVE LOGITS
     +=
    0.72
     |=
    0.69
     .=
    0.60
    ++;
    0.58
    addKill
    0.56
     isValid
    0.55
     mediaPlayer
    0.55
     добавля
    0.55
    respArray
    0.53
     ^=
    0.53
    Act Density 0.017%

    No Known Activations