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

    Neuronpedia

    INDEX
    Explanations

    describing specific states

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     ду
    0.52
     алге
    0.46
     перено
    0.44
     хотел
    0.43
     באמצעות
    0.43
     ער
    0.41
     보험
    0.41
     (\"
    0.41
    ুরী
    0.40
    嘣
    0.40
    POSITIVE LOGITS
     central
    0.48
     become
    0.46
    central
    0.44
     mar
    0.43
     explot
    0.43
     meta
    0.42
     memory
    0.42
    mio
    0.42
     randomly
    0.41
    زء
    0.41
    Act Density 0.030%

    No Known Activations