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

    Neuronpedia

    INDEX
    Explanations

    proper names

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     subtract
    0.57
     XYZ
    0.54
     DDD
    0.54
     BBB
    0.52
     GameObject
    0.52
      
    0.52
     dedu
    0.50
     whiteboard
    0.50
     xyz
    0.50
     usability
    0.49
    POSITIVE LOGITS
    lerin
    0.80
     était
    0.78
     tinha
    0.77
     aveva
    0.76
     zahlreichen
    0.76
    elyn
    0.75
     afirmó
    0.75
    baden
    0.74
     mostró
    0.73
     comenzó
    0.73
    Act Density 0.188%

    No Known Activations