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

    Neuronpedia

    INDEX
    Explanations

    penalties and premiums

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     zach
    -0.07
     DAY
    -0.07
     Day
    -0.07
     Bias
    -0.07
     initialization
    -0.07
     roses
    -0.07
     failure
    -0.07
     job
    -0.06
     věci
    -0.06
     Polly
    -0.06
    POSITIVE LOGITS
    Forge
    0.07
     withholding
    0.07
    ElementsBy
    0.06
     sanctions
    0.06
    OURCE
    0.06
    mayın
    0.06
    PIO
    0.06
    CellValue
    0.06
    MSG
    0.06
     '".
    0.06
    Act Density 0.101%

    No Known Activations