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

    Neuronpedia

    INDEX
    Explanations

    React code imports

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    RELATIVA
    0.79
     predictors
    0.78
    handedly
    0.76
    ⬥
    0.76
    れません
    0.72
     예측
    0.72
     problemy
    0.71
    】.
    0.71
     проблемы
    0.70
    ²/
    0.70
    POSITIVE LOGITS
    ,{
    0.98
    ,
    0.91
     Currie
    0.74
     almost
    0.74
    ięcy
    0.73
    ული
    0.71
    良
    0.71
     hampir
    0.65
     প্রায়
    0.64
     আকাঙ্
    0.64
    Act Density 0.003%

    No Known Activations