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

    Neuronpedia

    INDEX
    Explanations

    any specific details

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Eventually
    1.06
     uiteindelijk
    1.04
     결국
    0.98
    大多数
    0.94
    Sometimes
    0.93
    sometimes
    0.92
     eventually
    0.89
    全ての
    0.89
    越来越
    0.89
     అనేక
    0.88
    POSITIVE LOGITS
     relevant
    1.82
     noteworthy
    1.76
     pertinent
    1.64
     notable
    1.48
    relevant
    1.46
     any
    1.42
     relevante
    1.39
     Relevant
    1.39
     relevantes
    1.39
     applicable
    1.28
    Act Density 0.207%

    No Known Activations