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

    Neuronpedia

    INDEX
    Explanations

    need / necessary

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    if
    0.93
    שה
    0.89
    cannot
    0.86
    ^-
    0.83
                
    0.81
    unless
    0.80
     sukham
    0.80
    FIFA
    0.78
    Ment
    0.78
    Addo
    0.76
    POSITIVE LOGITS
    只需
    0.98
     intric
    0.91
     exclaimed
    0.90
     intuit
    0.88
     alfalfa
    0.88
    plemente
    0.88
     dàng
    0.85
    änger
    0.84
     টেক
    0.83
     glance
    0.82
    Act Density 0.087%

    No Known Activations