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

    Neuronpedia

    INDEX
    Explanations

    4

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     счита
    -0.07
    _validation
    -0.07
    凡
    -0.07
     cô
    -0.07
     Tas
    -0.07
    Candidate
    -0.07
    summary
    -0.06
    忱
    -0.06
     classmates
    -0.06
     soldiers
    -0.06
    POSITIVE LOGITS
     episodes
    0.07
    pile
    0.07
    \Message
    0.07
    extAlignment
    0.07
    のために
    0.07
     wakeup
    0.07
    轳
    0.07
    izzling
    0.07
    fork
    0.07
    *dt
    0.07
    Act Density 0.002%

    No Known Activations