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

    Neuronpedia

    INDEX
    Explanations

    set to some

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     оригіналу
    0.62
     следует
    0.61
    Alternatively
    0.61
    Become
    0.61
    才是
    0.60
    Ultimately
    0.60
    ProxyAgent
    0.58
    unless
    0.57
    ⇵
    0.57
     высоко
    0.57
    POSITIVE LOGITS
     några
    1.03
     some
    1.01
     bazı
    0.99
     quelques
    0.98
     néhány
    0.96
     alguns
    0.93
     alcuni
    0.93
     qualche
    0.92
     některé
    0.90
     vài
    0.89
    Act Density 1.079%

    No Known Activations