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

    Neuronpedia

    INDEX
    Explanations

    agent or source

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    data
    0.24
     data
    0.24
     simplified
    0.23
    化
    0.22
    データの
    0.22
     basic
    0.21
    making
    0.21
    disconnect
    0.21
     iteratively
    0.21
    数据
    0.20
    POSITIVE LOGITS
     someone
    0.38
     seseorang
    0.38
     একজন
    0.37
     whoever
    0.36
     alguien
    0.36
     anyone
    0.34
     iemand
    0.33
     alguém
    0.33
     jemand
    0.32
     him
    0.31
    Act Density 0.118%

    No Known Activations