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

    Neuronpedia

    INDEX
    Explanations

    unterteilt nach

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     long
    0.54
     preference
    0.54
     compos
    0.50
     whip
    0.49
     crowned
    0.49
    в
    0.49
     triple
    0.48
     hallmark
    0.48
     twofold
    0.47
     threefold
    0.47
    POSITIVE LOGITS
    ।’
    0.52
    AVLNode
    0.49
     కానీ
    0.48
    䉍
    0.46
    ’।
    0.46
    ꒪
    0.45
     ।
    0.44
     вступи
    0.44
    䐍
    0.44
    ”।
    0.44
    Act Density 0.000%

    No Known Activations