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

    Neuronpedia

    INDEX
    Explanations

    well versed

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    well
    0.98
    wells
    0.97
    WELL
    0.92
    Well
    0.91
     well
    0.85
     wel
    0.84
     oluşan
    0.82
     Well
    0.81
     crafted
    0.79
    存在する
    0.79
    POSITIVE LOGITS
     into
    0.84
     along
    0.74
    Su
    0.73
     beyond
    0.72
     Su
    0.72
    HEX
    0.71
    Hex
    0.71
    acqu
    0.71
     sui
    0.70
     ได้
    0.70
    Act Density 0.044%

    No Known Activations