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

    Neuronpedia

    INDEX
    Explanations

    construct

    np_max-act-logits · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     motion
    0.72
     Irwin
    0.69
     saucer
    0.68
    ಕರ್
    0.67
     നട
    0.66
     insignia
    0.66
     യാത്ര
    0.65
     motions
    0.65
    粉
    0.65
     panther
    0.64
    POSITIVE LOGITS
    construct
    0.94
    Construct
    0.87
    constructed
    0.74
     construct
    0.73
     konstru
    0.72
     Constructs
    0.70
     Construct
    0.69
    konstru
    0.68
    构造
    0.66
     এখ
    0.66
    Act Density 0.000%

    No Known Activations