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

    Neuronpedia

    INDEX
    Explanations

    root cause

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     feats
    2.76
     zabaw
    2.70
     حاجه
    2.63
    under
    2.59
     Suiza
    2.57
     Oiseaux
    2.52
    ಾಯಕ
    2.51
    ва
    2.51
     hurled
    2.50
    JLabel
    2.49
    POSITIVE LOGITS
    柢
    4.30
    お
    3.41
    ある
    2.69
    不算
    2.67
    𝖾
    2.61
    おい
    2.51
    𝑠
    2.50
     jaar
    2.49
    nię
    2.46
    ста
    2.43
    Act Density 0.049%

    No Known Activations