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

    Neuronpedia

    INDEX
    Explanations

    in advance

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     operative
    0.73
    oded
    0.68
    波动
    0.66
     oper
    0.64
     عم
    0.64
    顏
    0.61
    くら
    0.60
    uden
    0.59
     Paths
    0.59
     Vols
    0.59
    POSITIVE LOGITS
     give
    1.05
    Give
    0.97
    give
    0.95
    Say
    0.93
     say
    0.89
    say
    0.89
    Send
    0.87
    Adv
    0.84
     come
    0.84
     donnant
    0.84
    Act Density 0.003%

    No Known Activations