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

    Neuronpedia

    INDEX
    Explanations

    Scientific publications/patents

    np_max-act · gemini-2.0-flash

    The neuron activates on numeric tokens corresponding to publication years in reference citations.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     電
    -0.07
    ارد
    -0.07
    comed
    -0.07
    ortion
    -0.07
     foll
    -0.06
    jourd
    -0.06
    icates
    -0.06
     loro
    -0.06
    pip
    -0.06
     알
    -0.06
    POSITIVE LOGITS
     topLevel
    0.06
     nextState
    0.06
     wrapping
    0.06
     Hä
    0.06
    Shell
    0.06
    getInstance
    0.06
     refute
    0.06
     Giant
    0.06
    (Unknown
    0.06
     siguiente
    0.06
    Act Density 0.005%

    No Known Activations