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

    Neuronpedia

    INDEX
    Explanations

    Understanding complex descriptions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Pelham
    0.45
     Graz
    0.44
     Westchester
    0.42
     Working
    0.40
     Graham
    0.40
    inare
    0.40
     Granada
    0.39
     Glucose
    0.39
    Graz
    0.39
    Working
    0.38
    POSITIVE LOGITS
    ﹉
    0.41
    visao
    0.41
     modals
    0.38
    रेख
    0.38
    zien
    0.38
    bentuk
    0.38
    bitos
    0.37
     shapes
    0.37
    বিবি
    0.37
     pSig
    0.37
    Act Density 0.000%

    No Known Activations