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

    Neuronpedia

    INDEX
    Explanations

    instructions

    np_max-act · gemini-2.0-flash

    The neuron responds to floating-point number tokens (e.g. “4.0078125”), i.e. it flags decimal numeric metadata inserted in the text.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     derail
    -0.06
     призначення
    -0.06
    ограф
    -0.06
     pract
    -0.06
    /add
    -0.06
    .getRoot
    -0.06
    ाण
    -0.06
    plets
    -0.06
    .NotFound
    -0.06
     отп
    -0.06
    POSITIVE LOGITS
     giveaways
    0.07
     쉽
    0.07
     hostility
    0.07
     Egg
    0.07
     resentment
    0.07
    elig
    0.07
     Puppy
    0.07
    大學
    0.07
     Whitney
    0.07
     marzo
    0.07
    Act Density 0.015%

    No Known Activations