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

    Neuronpedia

    INDEX
    Explanations

    exiting loops

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     prettier
    0.42
    canoe
    0.41
     Его
    0.41
    ไข
    0.41
    σουμε
    0.39
    MatContext
    0.38
    творення
    0.38
    ༧
    0.38
    াক্ত
    0.38
    㚼
    0.38
    POSITIVE LOGITS
     resorts
    0.42
     patrones
    0.41
     otherwise
    0.40
     forse
    0.40
     sonst
    0.39
     suchen
    0.39
     Resort
    0.38
     Worst
    0.38
     continues
    0.38
     rings
    0.38
    Act Density 0.023%

    No Known Activations