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

    Neuronpedia

    INDEX
    Explanations

    this followed by a noun

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     proponents
    0.33
    etheless
    0.31
     bebidas
    0.31
    ate
    0.29
     rehearsals
    0.29
     gotas
    0.29
     always
    0.29
     toujours
    0.29
     plupart
    0.29
     pensare
    0.29
    POSITIVE LOGITS
    У
    0.37
    Съ
    0.37
    Ли
    0.34
    Си
    0.32
    Му
    0.32
    Ла
    0.32
    Ud
    0.31
    。
    0.31
    Су
    0.30
    Из
    0.30
    Act Density 0.089%

    No Known Activations