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

    Neuronpedia

    INDEX
    Explanations

    this + time period

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    et
    0.94
    و
    0.86
    ist
    0.75
    esque
    0.75
    o
    0.73
    ed
    0.71
     initiator
    0.71
    ing
    0.70
    us
    0.69
     juniper
    0.69
    POSITIVE LOGITS
    ра
    0.87
     связан
    0.77
    都可以
    0.74
    that
    0.73
    г
    0.73
    {
    0.71
    де
    0.70
    пи
    0.70
    дә
    0.70
    vý
    0.70
    Act Density 0.009%

    No Known Activations