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

    Neuronpedia

    INDEX
    Explanations

    Maintaining a constant state

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     prescription
    -0.07
     arcade
    -0.06
     Focus
    -0.06
     Connections
    -0.06
     Ctrl
    -0.06
     notices
    -0.06
    /provider
    -0.06
    finish
    -0.06
     deleted
    -0.06
    좌
    -0.06
    POSITIVE LOGITS
    <bits
    0.07
     وش
    0.07
    .backward
    0.07
    [date
    0.06
    .ReLU
    0.06
     LeBron
    0.06
    -tabs
    0.06
    英语
    0.06
     przez
    0.06
    nání
    0.06
    Act Density 0.058%

    No Known Activations