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

    Neuronpedia

    INDEX
    Explanations

    also

    np_max-act · gemini-2.0-flash

    The neuron detects discourse‐structuring cue words (like “One,” “Another,” “In addition,” “Overall”) that introduce or transition between list items or sections.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    .Next
    -0.07
     intrigued
    -0.07
    .infinity
    -0.06
     sector
    -0.06
    Moreover
    -0.06
    ![↵
    -0.06
    >")↵
    -0.06
     threatening
    -0.06
    热
    -0.06
    leftJoin
    -0.06
    POSITIVE LOGITS
     ItemStack
    0.06
     haciendo
    0.06
    512
    0.06
    指数
    0.06
     داخلی
    0.06
    _memcpy
    0.06
     файла
    0.06
    ssi
    0.06
    _GL
    0.06
    šil
    0.06
    Act Density 0.044%

    No Known Activations