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

    Neuronpedia

    INDEX
    Explanations

    hidden secrets

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ين
    0.58
    ции
    0.53
    ра
    0.50
    르
    0.49
     `:`,
    0.48
    ных
    0.48
    ва
    0.47
    ные
    0.46
    ラ
    0.46
     consci
    0.45
    POSITIVE LOGITS
    at
    0.76
    in
    0.64
    ↵
    0.64
    U
    0.63
    ur
    0.61
    7
    0.58
    Secret
    0.55
    AK
    0.55
    Secrets
    0.54
    UT
    0.53
    Act Density 0.365%

    No Known Activations