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

    Neuronpedia

    INDEX
    Explanations

    task or tasks

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ۳
    0.67
    ﺖ
    0.67
    тного
    0.65
    𝘵
    0.61
    argaan
    0.60
    ot
    0.59
    ្រ
    0.59
    3
    0.58
    트를
    0.56
    inę
    0.56
    POSITIVE LOGITS
    任务
    1.03
     tasks
    0.84
    ة
    0.84
    tasks
    0.83
    פ
    0.82
     tarefas
    0.81
     выпол
    0.80
     Aufgaben
    0.80
     Tasks
    0.79
     задачи
    0.78
    Act Density 0.061%

    No Known Activations