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

    Neuronpedia

    INDEX
    Explanations

    create, build, generate

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     honors
    0.68
    হিত
    0.67
     honours
    0.63
    स्पेक्ट
    0.62
     honoured
    0.62
    ienka
    0.62
    Match
    0.59
     rinsed
    0.58
     honored
    0.58
     intervened
    0.58
    POSITIVE LOGITS
     creation
    3.02
     create
    2.93
     creating
    2.91
     Creating
    2.88
     Create
    2.86
    create
    2.84
    Create
    2.81
     Creation
    2.81
    Creating
    2.80
     created
    2.76
    Act Density 3.616%

    No Known Activations