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

    Neuronpedia

    INDEX
    Explanations

    crystal meth

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    path
    0.51
    component
    0.50
    caused
    0.50
    propag
    0.50
    col
    0.49
    convex
    0.48
    config
    0.48
    could
    0.48
    was
    0.48
    becomes
    0.48
    POSITIVE LOGITS
     Waver
    0.51
    柺
    0.46
     Breaking
    0.45
    あなた
    0.45
     breaking
    0.44
     Nikki
    0.43
     привіт
    0.43
     admissions
    0.43
     y
    0.42
     Waiver
    0.42
    Act Density 0.001%

    No Known Activations