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

    Neuronpedia

    INDEX
    Explanations

    ever, all time

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    可以直接
    0.84
     closest
    0.77
     way
    0.77
     option
    0.76
     प्रकार
    0.73
     simplest
    0.73
     solutions
    0.73
    way
    0.71
     solution
    0.71
     options
    0.71
    POSITIVE LOGITS
     ever
    3.11
     EVER
    2.67
     Ever
    2.57
    Ever
    2.54
    ever
    2.44
     ooit
    2.21
    EVER
    2.16
     jamais
    1.71
     jamás
    1.65
     pernah
    1.51
    Act Density 0.062%

    No Known Activations