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

    Neuronpedia

    INDEX
    Explanations

    upstream/stream

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     Fixed
    -0.07
     proprietary
    -0.06
    RIA
    -0.06
     -------------------------------------------------------------------------↵
    -0.06
     различных
    -0.06
    ۴
    -0.06
    Fixed
    -0.06
     digits
    -0.05
    ,arr
    -0.05
     Пло
    -0.05
    POSITIVE LOGITS
    bel
    0.07
    vector
    0.07
    65
    0.07
     vybav
    0.07
    irected
    0.06
    graduate
    0.06
     Dai
    0.06
    ########################
    0.06
    _rl
    0.06
     GAL
    0.06
    Act Density 0.004%

    No Known Activations