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

    Neuronpedia

    INDEX
    Explanations

    Continuous Delivery

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     maximizes
    0.59
     can
    0.58
     recuer
    0.57
    ೃಷ್ಣ
    0.55
    Muchas
    0.55
     protects
    0.54
     spies
    0.53
     may
    0.53
    ગુ
    0.53
     prevented
    0.52
    POSITIVE LOGITS
     NYC
    0.61
     London
    0.59
    hā
    0.57
     On
    0.53
    instit
    0.53
    strate
    0.52
    ων
    0.52
    degree
    0.52
     domicile
    0.52
     Morgan
    0.51
    Act Density 0.000%

    No Known Activations