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

    Neuronpedia

    INDEX
    Explanations

    you are experiencing

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     hoera
    0.39
     Weekly
    0.38
     notification
    0.38
     grapevine
    0.38
     curiosity
    0.37
     buckling
    0.37
     novità
    0.37
     gravitate
    0.37
     detrás
    0.36
     curvatures
    0.36
    POSITIVE LOGITS
     think
    0.47
    std
    0.40
    think
    0.39
     THINK
    0.39
    หรือ
    0.39
    自身
    0.38
    ])[
    0.37
    have
    0.37
    および
    0.37
    0.36
    Act Density 0.002%

    No Known Activations