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

    Neuronpedia

    INDEX
    Explanations

    what follows

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ،
    0.22
    。
    0.21
    ，
    0.20
    、
    0.19
    "
    0.18
     prevails
    0.18
     there
    0.17
     podem
    0.17
    '
    0.17
     quiser
    0.17
    POSITIVE LOGITS
     allowing
    0.29
     culminating
    0.29
     necessitating
    0.29
     waardoor
    0.26
     waarbij
    0.26
     providing
    0.25
     featuring
    0.25
     જેમાં
    0.25
    allowing
    0.25
    从而
    0.25
    Act Density 0.259%

    No Known Activations