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

    Neuronpedia

    INDEX
    Explanations

    constrained math problems

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    erg
    -0.08
     escenario
    -0.08
     thereof
    -0.07
    udies
    -0.07
    opause
    -0.07
     homogen
    -0.07
    gorm
    -0.07
     powod
    -0.07
     qe
    -0.07
     dangere
    -0.07
    POSITIVE LOGITS
     нийл
    0.09
     naanị
    0.08
    Repe
    0.08
    andatu
    0.08
    Repeated
    0.08
     repetition
    0.08
     amháin
    0.08
    Resource
    0.08
    ’।
    0.08
     продукт
    0.08
    Act Density 0.006%

    No Known Activations