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

    Neuronpedia

    INDEX
    Explanations

    compatible

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    31
    -0.07
     drawing
    -0.07
    �
    -0.07
     vole
    -0.07
     Sell
    -0.07
     setters
    -0.07
     falling
    -0.06
    �
    -0.06
    156
    -0.06
    �
    -0.06
    POSITIVE LOGITS
     compatible
    0.16
     Compatible
    0.12
     compatibility
    0.12
    compatible
    0.10
    -compatible
    0.10
    Compatible
    0.10
    compat
    0.09
     Compatibility
    0.09
    _COMPAT
    0.09
     incompatible
    0.08
    Act Density 0.008%

    No Known Activations