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

    Neuronpedia

    INDEX
    Explanations

    variable name

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    learning
    0.51
    traded
    0.50
    deaths
    0.49
    importance
    0.48
    importanza
    0.48
    gangenheit
    0.46
     uptick
    0.46
    relevant
    0.46
    reise
    0.45
    teaching
    0.44
    POSITIVE LOGITS
     and
    0.43
     that
    0.41
     trabajar
    0.40
     Т
    0.40
     straighten
    0.40
     عنوان
    0.40
     works
    0.39
     ligation
    0.39
     arbejde
    0.39
    oglas
    0.39
    Act Density 0.002%

    No Known Activations