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

    Neuronpedia

    INDEX
    Explanations

    how not to

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    synt
    0.50
     synt
    0.48
    合成
    0.46
     sintet
    0.44
     synthesize
    0.44
    zenesulf
    0.44
     सूचित
    0.43
     synthesizing
    0.42
     sint
    0.42
     リン
    0.42
    POSITIVE LOGITS
    ↵
    0.50
     Еўропы
    0.46
    Planning
    0.45
    Cabin
    0.45
     Königreich
    0.44
    UCH
    0.44
    Sprint
    0.44
    établir
    0.44
    Fiat
    0.43
    Mars
    0.43
    Act Density 0.005%

    No Known Activations