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

    Neuronpedia

    INDEX
    Explanations

    Final

    np_max-act-logits · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ťou
    0.70
    ├──
    0.69
    सीय
    0.67
    送到
    0.67
    视为
    0.66
     رکھا
    0.66
     jaune
    0.65
    ymology
    0.65
    ENTRY
    0.65
     jasa
    0.64
    POSITIVE LOGITS
     before
    0.71
    前に
    0.68
     ar
    0.66
     スーツ
    0.66
     panel
    0.66
     risult
    0.66
    展示
    0.66
    pus
    0.64
     pants
    0.63
     tables
    0.63
    Act Density 0.005%

    No Known Activations