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

    Neuronpedia

    INDEX
    Explanations

    than doubled

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ه
    0.94
    ి
    0.93
    ে
    0.86
     Rhiz
    0.81
     Computed
    0.81
    퍄
    0.80
    эф
    0.79
    де
    0.77
     Yak
    0.76
     Stor
    0.76
    POSITIVE LOGITS
    ibar
    0.91
    away
    0.83
    icz
    0.83
    owaniem
    0.83
     divulg
    0.82
     dvou
    0.82
    iman
    0.80
    jedno
    0.80
    mäß
    0.79
    vents
    0.79
    Act Density 0.000%

    No Known Activations