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

    Neuronpedia

    INDEX
    Explanations

    main method using args

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    чной
    0.45
    nera
    0.42
    ၤ
    0.42
     vagin
    0.42
    需求的
    0.41
    Парт
    0.41
     bungalows
    0.40
     panne
    0.40
     lát
    0.40
    дование
    0.39
    POSITIVE LOGITS
    ti
    0.56
    itley
    0.51
    argv
    0.50
     argv
    0.46
    oud
    0.45
    if
    0.45
    aj
    0.44
    ators
    0.44
    ations
    0.43
    ㅋ
    0.43
    Act Density 0.003%

    No Known Activations