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

    Neuronpedia

    INDEX
    Explanations

    describing how things are

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    নকে
    0.78
     hiszen
    0.72
     совершен
    0.71
     निरंतर
    0.71
    mlp
    0.71
     celebration
    0.70
    无论是
    0.70
    maß
    0.69
     harmonious
    0.68
     осуществления
    0.67
    POSITIVE LOGITS
     usually
    1.06
    usually
    1.02
     meestal
    0.97
    instructions
    0.87
     biasanya
    0.82
     Usually
    0.82
    াতিক
    0.81
    把你
    0.80
     vary
    0.80
     varies
    0.79
    Act Density 0.775%

    No Known Activations