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

    Neuronpedia

    INDEX
    Explanations

    hidden secrets and troubles

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    pairwise
    0.70
     थिए
    0.67
     Wizards
    0.66
    covariance
    0.66
    ކ
    0.65
    다라고
    0.65
    |.|.|
    0.64
     especificar
    0.64
    😋
    0.63
    खन
    0.63
    POSITIVE LOGITS
     mysterious
    1.22
     secret
    1.13
     enigmatic
    1.12
     estranged
    1.11
     secrets
    1.07
     troubled
    1.06
     tragic
    1.02
     dysfunctional
    1.00
     unorthodox
    0.98
     disillusioned
    0.97
    Act Density 0.379%

    No Known Activations