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

    Neuronpedia

    INDEX
    Explanations

    connect

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    柟
    0.47
     ايش
    0.46
    一個
    0.45
     اين
    0.45
    ្នាក់
    0.43
     ذكر
    0.43
     cực
    0.42
     gehört
    0.42
     erwäh
    0.42
    بير
    0.41
    POSITIVE LOGITS
     Connect
    0.95
     connect
    0.90
     connects
    0.84
    Connect
    0.82
     Connecting
    0.82
     connecting
    0.81
     connecter
    0.76
     reconnect
    0.69
     CONNECT
    0.69
    コネ
    0.69
    Act Density 0.024%

    No Known Activations