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

    Neuronpedia

    INDEX
    Explanations

    say "the ending -ose"

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     gör
    -0.08
    堨
    -0.07
     antid
    -0.07
     Joan
    -0.07
    🔵
    -0.07
    橐
    -0.07
    .Raycast
    -0.07
    짥
    -0.07
    inaire
    -0.07
    .clearRect
    -0.07
    POSITIVE LOGITS
    Style
    0.07
    情
    0.07
    ству
    0.07
    슬
    0.06
    ategy
    0.06
    unity
    0.06
    .lu
    0.06
     occupation
    0.06
    สล
    0.06
    класс
    0.06
    Act Density 0.002%

    No Known Activations