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

    Neuronpedia

    INDEX
    Explanations

    will be and future plans

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     intervention
    0.77
    how
    0.68
     versions
    0.66
    How
    0.66
     desirable
    0.65
     dés
    0.64
     assent
    0.64
     পছ
    0.64
     subsequence
    0.64
    desired
    0.63
    POSITIVE LOGITS
     inoltre
    0.86
     głównie
    0.84
     även
    0.80
    பட்ச
    0.78
     hvert
    0.77
     প্রতিটি
    0.77
     mainly
    0.77
     primarily
    0.74
     estará
    0.73
     sicuramente
    0.73
    Act Density 0.159%

    No Known Activations