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

    Neuronpedia

    INDEX
    Explanations

    combination of

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     aromat
    0.47
     frictionless
    0.45
     COVID
    0.44
     toler
    0.43
    為什麼
    0.42
    ’!
    0.42
     chronically
    0.42
     Covid
    0.41
     mastectomy
    0.41
     bullshit
    0.41
    POSITIVE LOGITS
    d
    0.59
    y
    0.51
    най
    0.49
    ęg
    0.47
    son
    0.47
     Además
    0.46
    Message
    0.46
    g
    0.46
    z
    0.46
    b
    0.44
    Act Density 0.004%

    No Known Activations