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

    Neuronpedia

    INDEX
    Explanations

    potential resources or entities

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    u
    0.53
    se
    0.47
    veter
    0.46
    ze
    0.46
    oriented
    0.44
    primarily
    0.44
    centric
    0.44
     consumes
    0.44
     primarily
    0.43
    layered
    0.43
    POSITIVE LOGITS
     thử
    0.49
     ق
    0.48
     eggs
    0.46
     magic
    0.46
     flavour
    0.45
     غ
    0.44
     xạ
    0.44
     brownies
    0.44
    惊喜
    0.44
     kiến
    0.44
    Act Density 0.004%

    No Known Activations