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

    Neuronpedia

    INDEX
    Explanations

    code coco

    np_max-act-logits · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     vertical
    0.91
     lifts
    0.84
     lifting
    0.83
     upright
    0.82
    Vertical
    0.81
     lift
    0.80
    垂直
    0.79
     lifted
    0.77
     verticales
    0.77
     вертика
    0.76
    POSITIVE LOGITS
     codes
    0.89
     Code
    0.85
     Codes
    0.83
     code
    0.75
    Code
    0.75
     Coco
    0.73
    codes
    0.71
     Light
    0.70
     códigos
    0.69
     coco
    0.66
    Act Density 0.219%

    No Known Activations