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

    Neuronpedia

    INDEX
    Explanations

    closing punctuation and quotes

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     అను
    0.80
     Ε
    0.77
    धित
    0.72
    দ্দীন
    0.69
     των
    0.68
     meng
    0.68
     Power
    0.67
     wind
    0.67
     Ether
    0.67
     entree
    0.66
    POSITIVE LOGITS
    ’:
    0.83
     “”
    0.80
    ”:
    0.79
    मारा
    0.79
     ”
    0.78
     că
    0.77
     ,,
    0.75
     ,"
    0.74
    effici
    0.73
    設け
    0.73
    Act Density 0.089%

    No Known Activations