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

    Neuronpedia

    INDEX
    Explanations

    method for

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    s
    0.65
    ﺓ
    0.64
    ，
    0.60
    。”
    0.57
    ে
    0.57
     '
    0.55
    ，「
    0.55
    ı
    0.55
    ร
    0.55
    ins
    0.54
    POSITIVE LOGITS
    dır
    0.68
     такое
    0.62
     HANDLING
    0.58
    inafter
    0.57
     écrite
    0.57
     associée
    0.55
     länger
    0.53
    ರ್ಚ
    0.52
     Alguns
    0.52
     ferrugineux
    0.52
    Act Density 3.417%

    No Known Activations