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

    Neuronpedia

    INDEX
    Explanations

    words related to reports and recommendations in data and research studies

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Negative Logits
    conservancy
    -1.01
    cffff
    -0.94
    antha
    -0.85
    yg
    -0.85
     hypert
    -0.82
    east
    -0.82
     ILCS
    -0.79
     municip
    -0.77
     appre
    -0.76
     bene
    -0.75
    POSITIVE LOGITS
    velop
    1.14
    iry
    1.07
     titled
    0.94
    recomm
    0.89
    emi
    0.89
    backs
    0.88
    cases
    0.86
    zes
    0.85
    marks
    0.84
    arian
    0.83
    Act Density 1.084%

    No Known Activations