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

    Neuronpedia

    INDEX
    Explanations

    graph and diagram generation

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     pist
    0.44
     бед
    0.44
    pist
    0.42
    〰
    0.40
    odic
    0.40
     piston
    0.39
     bø
    0.38
     loin
    0.38
     Pist
    0.38
     piste
    0.37
    POSITIVE LOGITS
     graph
    0.92
     Graph
    0.89
     Graphs
    0.80
    Graph
    0.79
     graphs
    0.75
     subgraph
    0.75
    graph
    0.71
    Graphs
    0.71
     Diagram
    0.68
    graphs
    0.67
    Act Density 0.016%

    No Known Activations