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

    Neuronpedia

    INDEX
    Explanations

    break down or apart

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     ಭಯ
    0.78
     এলি
    0.77
    oğlu
    0.73
    пле
    0.71
     ટકા
    0.70
    pct
    0.69
    ppin
    0.69
    ร่วม
    0.68
    ufs
    0.68
     своє
    0.68
    POSITIVE LOGITS
     apart
    1.72
     Apart
    1.49
    apart
    1.46
    Apart
    1.41
     down
    1.35
    down
    1.30
    Down
    1.14
     Down
    1.14
     into
    1.13
    hearted
    1.12
    Act Density 0.149%

    No Known Activations