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

    Neuronpedia

    INDEX
    Explanations

    thin film or layer

    np_max-act-logits · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Step
    0.95
    step
    0.87
     step
    0.86
    Stage
    0.85
    stage
    0.84
     stage
    0.79
    단계
    0.78
    Lu
    0.75
     layer
    0.75
    বর
    0.74
    POSITIVE LOGITS
     sintered
    0.79
     women
    0.70
     Clothing
    0.70
     painters
    0.66
     Women
    0.66
     travellers
    0.65
    ര്‍ഷ
    0.65
     mammals
    0.65
     &$\
    0.65
     женщины
    0.65
    Act Density 0.005%

    No Known Activations