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

    Neuronpedia

    INDEX
    Explanations

    period

    np_max-act · gemini-2.0-flash

    The neuron detects discourse-marker or transitional words and phrases (e.g. “for example,” “while,” “if,” “only,” etc.) that signal shifts, contrasts, or examples in the text.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     tightening
    -0.06
     místo
    -0.06
     Т
    -0.06
    ihu
    -0.05
    ाइम
    -0.05
     дан
    -0.05
    人
    -0.05
    비스
    -0.05
    .magic
    -0.05
    両
    -0.05
    POSITIVE LOGITS
    .urlencoded
    0.07
    .bg
    0.07
    _CLEAR
    0.06
     существ
    0.06
    ops
    0.06
     tecn
    0.06
    transpose
    0.06
     něm
    0.06
    '],↵
    0.06
    878
    0.06
    Act Density 0.227%

    No Known Activations