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

    Neuronpedia

    INDEX
    Explanations

    persistent

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     Stamford
    -0.06
    .Tree
    -0.06
    .Subject
    -0.06
    ＾
    -0.06
    .Measure
    -0.06
    -Co
    -0.06
     blinds
    -0.06
     assum
    -0.06
     combin
    -0.06
    uled
    -0.06
    POSITIVE LOGITS
     persistence
    0.08
    .persistence
    0.07
     Ped
    0.07
     Tara
    0.07
     persisted
    0.07
     persistent
    0.07
     ghost
    0.07
     hdf
    0.06
    _rng
    0.06
    ifications
    0.06
    Act Density 0.003%

    No Known Activations