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

    Neuronpedia

    INDEX
    Explanations

    potential errors or not standard

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    WinCounter
    0.42
    PTMR
    0.41
     Verantwort
    0.39
     mitigating
    0.38
    Career
    0.38
    achtet
    0.36
    रिक
    0.36
     TGC
    0.36
    कीस
    0.35
    מו
    0.35
    POSITIVE LOGITS
     gen
    0.43
    wan
    0.40
    Located
    0.40
     refer
    0.39
     located
    0.39
     resonate
    0.38
    xl
    0.38
    located
    0.38
    ostr
    0.38
    nienie
    0.37
    Act Density 0.000%

    No Known Activations