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

    Neuronpedia

    INDEX
    Explanations

    and, but, with

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    took
    0.45
    ются
    0.44
     enables
    0.44
     může
    0.43
     inhibits
    0.43
     motivates
    0.43
    шает
    0.43
    한다
    0.43
     없다
    0.43
     decides
    0.42
    POSITIVE LOGITS
     featuring
    1.31
     requiring
    1.11
    featuring
    1.02
     pouvant
    1.01
     containing
    0.99
     consisting
    0.99
     Featuring
    0.99
     needing
    0.98
     possessing
    0.98
     comprising
    0.94
    Act Density 0.011%

    No Known Activations