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

    Neuronpedia

    INDEX
    Explanations

    main method args

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Argent
    0.48
    []=
    0.45
     Argentine
    0.43
     Salas
    0.42
    ◻
    0.41
    Argentina
    0.40
     Argentina
    0.40
    Agence
    0.40
     Cortes
    0.39
    remaining
    0.39
    POSITIVE LOGITS
    数为
    0.36
    िट
    0.34
     uj
    0.34
    ücht
    0.34
    ጄ
    0.34
     Это
    0.34
     trận
    0.34
    গ্
    0.34
    辩
    0.33
     gerektiğini
    0.33
    Act Density 0.001%

    No Known Activations