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

    Neuronpedia

    INDEX
    Explanations

    Whenever, Always, Before

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    い
    0.45
    ی
    0.43
     хотя
    0.42
     באי
    0.40
    琹
    0.40
    ים
    0.40
    NuGet
    0.39
    答应
    0.39
    რ
    0.39
     специалист
    0.39
    POSITIVE LOGITS
     Whenever
    0.44
     Formerly
    0.42
     See
    0.40
     BEFORE
    0.40
     Burn
    0.40
     Sky
    0.39
     Always
    0.38
     Anytime
    0.38
     hub
    0.38
     Elev
    0.38
    Act Density 0.005%

    No Known Activations