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

    Neuronpedia

    INDEX
    Explanations

    word parts

    np_max-act-logits · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ertid
    0.38
    lein
    0.38
     dares
    0.36
     esse
    0.36
    conc
    0.36
     trac
    0.35
    apuram
    0.35
    篆
    0.35
    agre
    0.34
    लास
    0.34
    POSITIVE LOGITS
     né
    0.44
    spent
    0.39
     hurled
    0.37
    abb
    0.37
    ABB
    0.37
     Spent
    0.37
    ყო
    0.36
     Desember
    0.36
     instructed
    0.36
     tossed
    0.36
    Act Density 0.000%

    No Known Activations