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

    Neuronpedia

    INDEX
    Explanations

    deliverypriorityexperiencesblossomingtop

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    New
    0.88
    荕
    0.86
    Get
    0.86
    Fishing
    0.85
    Media
    0.85
    Pe
    0.83
    Participant
    0.82
     rotacao
    0.81
    Grain
    0.81
    Change
    0.80
    POSITIVE LOGITS
     👇
    0.81
    <0xAD>
    0.80
     but
    0.76
     equaling
    0.71
     seems
    0.70
     ting
    0.69
     👌
    0.68
     diminishes
    0.68
     seem
    0.68
    <0xAC>
    0.67
    Act Density 0.374%

    No Known Activations