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

    Neuronpedia

    INDEX
    Explanations

    Rest allows/the Internet/attribution

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     pubb
    0.78
    ൧
    0.78
     emphasised
    0.76
     detal
    0.73
    تور
    0.73
     হেসে
    0.73
     exce
    0.70
    ისტ
    0.70
     offen
    0.70
    开始
    0.70
    POSITIVE LOGITS
     इन्वेस्ट
    0.89
     stylu
    0.86
    ಾಸ
    0.85
     உற
    0.85
     Ansel
    0.83
     assignee
    0.82
     Steel
    0.82
     Cig
    0.82
     ആരോപ
    0.81
     Riders
    0.81
    Act Density 0.001%

    No Known Activations