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

    Neuronpedia

    INDEX
    Explanations

    ]

    np_max-act-logits · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     ②
    0.73
     ④
    0.68
     suburbs
    0.67
     suburban
    0.67
     ①
    0.67
     Doesn
    0.67
     lantern
    0.67
     ‘
    0.65
     Against
    0.64
     ——
    0.64
    POSITIVE LOGITS
    Hos
    0.68
     Collin
    0.67
    ப்படுகிறது
    0.65
    жни
    0.61
    shear
    0.60
    obs
    0.59
    כים
    0.59
    垚
    0.59
    销
    0.59
    খ্য
    0.59
    Act Density 0.017%

    No Known Activations