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

    Neuronpedia

    INDEX
    Explanations

    cardinal directions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     call
    0.76
    每一个
    0.75
     每个
    0.74
    毎回
    0.73
     ویژگی
    0.71
     juggle
    0.70
     ENV
    0.70
     주는
    0.69
     susp
    0.68
     환경
    0.68
    POSITIVE LOGITS
     east
    1.86
     west
    1.85
     north
    1.83
     south
    1.82
     northeast
    1.71
    north
    1.66
     northward
    1.66
     eastward
    1.66
    east
    1.61
     eastern
    1.61
    Act Density 0.284%

    No Known Activations