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

    Neuronpedia

    INDEX
    Explanations

    loading web pages and models

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     mistakes
    0.45
     leaks
    0.44
     Mig
    0.42
     headaches
    0.41
     illnesses
    0.41
    पढ़
    0.40
    એ
    0.39
     spikes
    0.39
     misunderstand
    0.38
     migraines
    0.38
    POSITIVE LOGITS
     page
    0.55
    page
    0.50
     chunk
    0.46
     webpage
    0.45
     iframe
    0.45
     intravenously
    0.45
    section
    0.44
     страницу
    0.44
    svg
    0.43
     img
    0.43
    Act Density 0.001%

    No Known Activations