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

    Neuronpedia

    INDEX
    Explanations

    paragraph without numbered format

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     heartfelt
    0.49
     shockingly
    0.49
     glaz
    0.49
     pretzel
    0.48
     shorthand
    0.48
     heartbreaking
    0.47
     dishon
    0.46
     već
    0.45
    izacji
    0.45
    6
    0.45
    POSITIVE LOGITS
    t
    0.56
    وضع
    0.52
    tolerance
    0.49
    hesda
    0.48
    ase
    0.46
    s
    0.46
    tra
    0.45
    seg
    0.45
    sop
    0.45
    hyd
    0.45
    Act Density 0.002%

    No Known Activations