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

    Neuronpedia

    INDEX
    Explanations

    listing points or ideas

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    伌
    0.46
    enskap
    0.42
    R
    0.41
    atorias
    0.38
     Размер
    0.38
    reate
    0.37
    んと
    0.37
    可以选择
    0.37
    াযোগ
    0.36
     necesit
    0.36
    POSITIVE LOGITS
     convergence
    0.47
    пла
    0.43
     converge
    0.42
     લખ
    0.42
     coalescence
    0.42
     artich
    0.41
    хождение
    0.41
     osse
    0.40
    više
    0.40
     concatenation
    0.39
    Act Density 0.002%

    No Known Activations