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

    Neuronpedia

    INDEX
    Explanations

    boosting rounds, smoking herbs, code structure definitions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     horizont
    0.86
    ยนตร์
    0.86
     Belinda
    0.84
     grill
    0.83
    ᖕ
    0.83
    顿
    0.82
     gras
    0.81
     stal
    0.81
     Redirect
    0.80
    蓥
    0.80
    POSITIVE LOGITS
    at
    1.05
    Ti
    1.04
    א
    1.04
    ק
    1.02
    r
    0.98
    ii
    0.97
    d
    0.95
    okane
    0.95
    은
    0.95
    на
    0.93
    Act Density 0.002%

    No Known Activations