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

    Neuronpedia

    INDEX
    Explanations

    this year's

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     These
    -2.30
     Both
    -1.76
     or
    -1.76
     for
    -1.75
     that
    -1.72
     noted
    -1.63
    年は
    -1.59
    飚
    -1.57
     Beide
    -1.56
     same
    -1.55
    POSITIVE LOGITS
     anhänger
    1.54
    boven
    1.52
     funcione
    1.51
     AspNetCore
    1.51
    ously
    1.51
    ͞
    1.48
     merke
    1.48
     российского
    1.46
     familières
    1.45
     え
    1.45
    Act Density 0.038%

    No Known Activations