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

    Neuronpedia

    INDEX
    Explanations

    period

    np_max-act · gemini-2.0-flash

    This neuron flags the boundaries of questions or sentences, especially the word “What” at the start of a question and the final period token.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     chocol
    -0.07
     rotor
    -0.06
    ，此
    -0.06
    Driver
    -0.06
     Однак
    -0.06
     retr
    -0.06
     age
    -0.06
     ServiceProvider
    -0.06
    œur
    -0.06
    �
    -0.06
    POSITIVE LOGITS
    buff
    0.07
    ी.
    0.07
    ーク
    0.06
    ά
    0.06
     Millions
    0.06
    studio
    0.06
    .Is
    0.06
    ViewSet
    0.06
     america
    0.06
    ая
    0.06
    Act Density 0.003%

    No Known Activations