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

    Neuronpedia

    INDEX
    Explanations

    Ag followed by grid, gender, or aggreg

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    eszcze
    0.48
     annat
    0.41
     كس
    0.39
     orta
    0.39
    ामुळे
    0.39
    hade
    0.37
    gahan
    0.37
    genden
    0.36
    gehend
    0.36
    oi
    0.36
    POSITIVE LOGITS
    Ag
    0.78
     Agg
    0.77
     agg
    0.73
     Ag
    0.71
     ag
    0.69
     aggregated
    0.66
     агре
    0.66
    Agg
    0.66
     aggregation
    0.65
     AG
    0.64
    Act Density 0.022%

    No Known Activations