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

    Neuronpedia

    INDEX
    Explanations

    code and programming namespaces

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    拍了拍
    -0.98
     Chateau
    -0.95
    Szia
    -0.93
    ⎊
    -0.93
     Secrétaire
    -0.91
     secretariat
    -0.91
     fön
    -0.90
     quelcon
    -0.89
     chateau
    -0.88
     tertarik
    -0.88
    POSITIVE LOGITS
     rus
    0.89
     whole
    0.81
    ます
    0.81
    些
    0.79
     savu
    0.79
     comple
    0.76
     bēr
    0.74
      
    0.73
     trap
    0.73
     extern
    0.73
    Act Density 0.001%

    No Known Activations