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

    Neuronpedia

    INDEX
    Explanations

    Citations

    np_max-act · gemini-2.0-flash

    This neuron responds to special control and metadata tokens that delimit and annotate parts of the conversation (e.g. start/end markers, header IDs, speaker tags).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    Paren
    -0.06
     Registr
    -0.06
     культур
    -0.06
    .pth
    -0.06
     такие
    -0.06
    HAV
    -0.06
     polít
    -0.06
    MBED
    -0.06
    魔法
    -0.06
    DEPEND
    -0.06
    POSITIVE LOGITS
     confident
    0.07
    Cond
    0.07
     스
    0.07
    ？
    0.07
    lds
    0.06
    .She
    0.06
     груз
    0.06
    .controller
    0.06
    .isPlaying
    0.06
     зер
    0.06
    Act Density 0.010%

    No Known Activations