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

    Neuronpedia

    INDEX
    Explanations

    MIME types

    np_max-act-logits · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Subgraph
    0.68
     cứng
    0.65
    🌖
    0.63
     Hfn
    0.62
    Seal
    0.61
    سٹر
    0.60
    Robinson
    0.60
     hick
    0.58
     taala
    0.58
     Hepat
    0.58
    POSITIVE LOGITS
     vy
    0.59
     reduction
    0.57
     منها
    0.52
     CR
    0.51
     typing
    0.49
    減少
    0.48
    ชา
    0.48
     Wes
    0.48
     представля
    0.47
     typology
    0.47
    Act Density 0.005%

    No Known Activations