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

    Neuronpedia

    INDEX
    Explanations
    No Explanations Found
    New Auto-Interp
    Negative Logits
     outweighs
    0.91
     decoupling
    0.87
     other
    0.87
     plus
    0.86
     sided
    0.84
     mis
    0.83
     triangulation
    0.82
     comparable
    0.81
     tetromino
    0.81
     tripartite
    0.80
    POSITIVE LOGITS
    Here
    1.98
    ##
    1.92
    Welcome
    1.79
    Dear
    1.71
    ```
    1.64
    Hello
    1.61
    Okay
    1.61
    Introduction
    1.60
    Below
    1.59
    Let
    1.55
    Act Density 1.737%

    No Known Activations