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

    Neuronpedia

    INDEX
    Explanations

    Code

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    _skills
    -0.07
     которое
    -0.07
    什
    -0.06
     hora
    -0.06
    Jo
    -0.06
    lbrakk
    -0.06
    flu
    -0.06
    關
    -0.06
     algo
    -0.06
     pity
    -0.06
    POSITIVE LOGITS
     Fel
    0.07
    _GF
    0.06
     Slee
    0.06
     объект
    0.06
     massage
    0.06
     disgr
    0.06
    /wp
    0.06
    örper
    0.06
     getString
    0.06
     embroid
    0.06
    Act Density 0.015%

    No Known Activations