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

    Neuronpedia

    INDEX
    Explanations

    turn

    np_max-act-logits · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     traffic
    0.86
     folders
    0.80
     Traffic
    0.79
    traffic
    0.79
     packers
    0.74
    Traffic
    0.72
     Airport
    0.70
    Folder
    0.69
    hurst
    0.69
     Scaffold
    0.69
    POSITIVE LOGITS
     turning
    1.07
     turn
    0.94
     Turning
    0.91
    turn
    0.90
    turning
    0.89
    Turning
    0.85
     Turn
    0.83
    Turn
    0.82
     turned
    0.80
     turns
    0.77
    Act Density 0.053%

    No Known Activations