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

    Neuronpedia

    INDEX
    Explanations

    specific case or context

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    countries
    0.45
     nella
    0.43
    including
    0.42
    是在
    0.41
    7
    0.40
    哪
    0.40
    btn
    0.40
    😲
    0.40
    在我
    0.39
    ՛
    0.39
    POSITIVE LOGITS
     regard
    0.75
     regards
    0.65
     instance
    0.61
     case
    0.59
     경우
    0.59
     случа
    0.53
     respect
    0.52
     caso
    0.51
     मामले
    0.51
     случае
    0.50
    Act Density 0.007%

    No Known Activations