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

    Neuronpedia

    INDEX
    Explanations

    Software weaknesses

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    torie
    -0.57
     labd
    -0.49
     labelling
    -0.48
     experimenter
    -0.48
    Total
    -0.47
    出版年
    -0.47
    gull
    -0.47
     analogue
    -0.47
     fallo
    -0.46
    Figure
    -0.46
    POSITIVE LOGITS
     There
    0.75
     Though
    0.69
     Thanks
    0.69
     Many
    0.66
     Unless
    0.65
     If
    0.65
     Although
    0.65
     Here
    0.64
     The
    0.64
     We
    0.64
    Act Density 0.040%

    No Known Activations