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

    Neuronpedia

    INDEX
    Explanations

    worn, damaged, or aged objects

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     亻
    0.36
    一定的
    0.36
     appreciable
    0.35
     Importantly
    0.35
     conducive
    0.35
    <unused2130>
    0.35
    嶠
    0.34
    <unused2164>
    0.34
     workmanship
    0.34
     belangrijk
    0.34
    POSITIVE LOGITS
     drunken
    0.62
     sweaty
    0.60
     dusty
    0.60
     gritty
    0.58
     soggy
    0.58
     greasy
    0.57
     rusty
    0.57
     jagged
    0.57
     cynical
    0.56
     rusted
    0.56
    Act Density 0.062%

    No Known Activations