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

    Neuronpedia

    INDEX
    Explanations

    items numbered 3

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Enter
    0.93
     forwarding
    0.85
     gian
    0.85
     parade
    0.84
     tw
    0.83
     equation
    0.83
     enter
    0.82
     labeling
    0.79
     lasers
    0.79
     narrative
    0.78
    POSITIVE LOGITS
    ه
    0.99
    even
    0.89
    Producto
    0.81
    Casual
    0.76
    Practical
    0.75
    casual
    0.75
    o
    0.74
    edad
    0.73
    غانستان
    0.73
    وال
    0.73
    Act Density 0.035%

    No Known Activations