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

    Neuronpedia

    INDEX
    Explanations

    data list and location

    np_max-act-logits · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Rachel
    0.97
    Diane
    0.93
     Lars
    0.89
    Harry
    0.87
     Lionel
    0.86
    Rachel
    0.85
    Martha
    0.84
     Beatrice
    0.84
    Robert
    0.84
    Patrick
    0.83
    POSITIVE LOGITS
    ാന്‍
    0.82
     list
    0.77
    ਾਬ
    0.75
     FOD
    0.74
    燥
    0.73
     d
    0.73
     socalled
    0.72
     unlucky
    0.72
     Really
    0.72
    uffling
    0.72
    Act Density 0.000%

    No Known Activations