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

    Neuronpedia

    INDEX
    Explanations

    tube, pipeline

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     Tube
    -1.35
     tube
    -1.31
    Tube
    -1.20
    tube
    -1.13
     Tubes
    -1.09
     pipeline
    -1.06
     tubes
    -1.05
     TUBE
    -1.05
    tubes
    -1.01
     Pipeline
    -0.93
    POSITIVE LOGITS
     defaultstate
    0.74
     admixture
    0.69
    <bos>
    0.69
    featureID
    0.69
    ConstraintMaker
    0.66
     perfons
    0.65
     anvil
    0.65
     ***!
    0.63
     potholes
    0.63
    contentLoaded
    0.62
    Act Density 0.173%

    No Known Activations