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

    Neuronpedia

    INDEX
    Explanations

    Data overwriting and recovery

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     unico
    -0.09
     tru
    -0.08
    modus
    -0.08
     herkennen
    -0.08
     trud
    -0.08
     coneg
    -0.08
     ganho
    -0.07
     everywhere
    -0.07
     únicos
    -0.07
     único
    -0.07
    POSITIVE LOGITS
     subsequent
    0.12
     subsequently
    0.10
     Subse
    0.10
     Subsequently
    0.09
     afterward
    0.09
     позже
    0.09
     interven
    0.09
     posteriormente
    0.09
     afterwards
    0.09
     thereafter
    0.09
    Act Density 0.040%

    No Known Activations