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

    Neuronpedia

    INDEX
    Explanations

    when done or reached

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     blieb
    0.40
     никогда
    0.40
     помогают
    0.40
     помога
    0.39
     često
    0.39
     поможет
    0.38
     ভুল
    0.37
     niemals
    0.37
    yscrapers
    0.36
     помогут
    0.36
    POSITIVE LOGITS
     selesai
    0.66
     finished
    0.59
     완료
    0.57
     xong
    0.56
    完成
    0.55
     completed
    0.54
     reaches
    0.53
     完成
    0.52
     completion
    0.51
     готов
    0.48
    Act Density 0.168%

    No Known Activations