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

    Neuronpedia

    INDEX
    Explanations

    in a phase or term

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    差点
    0.42
     যেদিন
    0.41
    那天
    0.41
    进步
    0.40
    成为
    0.39
     όταν
    0.36
     অনুষ্ঠানে
    0.36
     maliciously
    0.36
     quando
    0.35
    isce
    0.35
    POSITIVE LOGITS
     phase
    0.76
     preparations
    0.73
     fase
    0.71
     Phase
    0.69
     phases
    0.62
     segundo
    0.62
    phase
    0.61
     proceso
    0.61
     second
    0.61
     দ্বিতীয়
    0.59
    Act Density 0.013%

    No Known Activations