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

    Neuronpedia

    INDEX
    Explanations

    CosmicQbit, ALiBi

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     horrified
    0.51
     얘는
    0.49
     soar
    0.49
     ailment
    0.47
    ﮏ
    0.47
     وهو
    0.46
     Aviso
    0.45
     և
    0.45
     frenzy
    0.45
     horrifying
    0.45
    POSITIVE LOGITS
    '
    0.76
    .
    0.59
    Than
    0.57
    ‘
    0.54
    |
    0.53
    ${
    0.52
    ForThe
    0.52
    Diamonds
    0.50
    In
    0.50
    на
    0.50
    Act Density 0.520%

    No Known Activations