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

    Neuronpedia

    INDEX
    Explanations

    module and let statements

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Cred
    0.78
     Lieferung
    0.77
     Reviewed
    0.76
     connaître
    0.74
    COUNTS
    0.73
     Info
    0.72
     Know
    0.71
    ურად
    0.71
    spann
    0.71
     originals
    0.70
    POSITIVE LOGITS
    วย
    0.63
     서로
    0.59
    ləşdir
    0.58
    ستان
    0.57
     billions
    0.57
     suppressed
    0.55
     doi
    0.55
     interrupts
    0.54
    dio
    0.52
     fork
    0.52
    Act Density 0.002%

    No Known Activations