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

    Neuronpedia

    INDEX
    Explanations

    activities followed by articles or objects

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    3
    1.21
    7
    1.10
    2
    1.00
    1
    0.95
    ת
    0.95
    is
    0.93
    5
    0.92
    9
    0.91
    6
    0.91
    al
    0.88
    POSITIVE LOGITS
     nécessairement
    0.88
     préférence
    0.84
     parfaite
    0.83
     considéré
    0.83
     dériv
    0.79
     rédu
    0.79
    ங்கிணை
    0.79
     teve
    0.78
    '
    0.78
     connu
    0.77
    Act Density 0.329%

    No Known Activations