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

    Neuronpedia

    INDEX
    Explanations

    sides and directions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     плен
    0.59
    inkle
    0.57
    ink
    0.57
    已有
    0.56
     nhân
    0.56
    สำเร็จ
    0.55
     mera
    0.55
     stages
    0.55
     χώρα
    0.55
     megsz
    0.53
    POSITIVE LOGITS
     left
    2.37
    left
    2.23
    Left
    2.18
     Left
    2.17
     LEFT
    1.90
    左
    1.86
    右
    1.83
     right
    1.82
     Right
    1.70
     izquierda
    1.69
    Act Density 0.567%

    No Known Activations