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

    Neuronpedia

    INDEX
    Explanations

    thought, through, out

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    .St
    -0.07
    -Ф
    -0.07
    784
    -0.07
     rule
    -0.06
    afort
    -0.06
     trim
    -0.06
    rep
    -0.06
    _WINDOW
    -0.06
    OD
    -0.06
    .lib
    -0.06
    POSITIVE LOGITS
     thought
    0.07
    thought
    0.06
    /close
    0.06
     pá
    0.06
    ans
    0.06
     %@",
    0.06
     rethink
    0.06
    ernity
    0.06
     Chiefs
    0.06
     країни
    0.06
    Act Density 0.015%

    No Known Activations