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

    Neuronpedia

    INDEX
    Explanations

    open-weights model

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    퀘
    0.44
     সহায়
    0.40
    quat
    0.38
    ım
    0.37
     ड्यू
    0.37
    定める
    0.37
    平方
    0.37
     нужда
    0.37
    हाय
    0.36
    再说
    0.36
    POSITIVE LOGITS
     details
    0.39
     identical
    0.38
     garnered
    0.37
    endeu
    0.36
     assets
    0.36
    ="#"
    0.35
     collar
    0.35
     маршру
    0.35
    ica
    0.35
     mo
    0.34
    Act Density 0.006%

    No Known Activations