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

    Neuronpedia

    INDEX
    Explanations

    questions

    np_max-act · gemini-2.0-flash

    The neuron is picking up common connective words—especially conjunctions (e.g. “if,” “or,” “and”) that link clauses.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    discussion
    -0.07
     이름
    -0.07
    ithe
    -0.07
    _wire
    -0.07
     pager
    -0.06
    StringBuilder
    -0.06
     editions
    -0.06
    //--------------------------------
    -0.06
    _instructions
    -0.06
    CGRect
    -0.06
    POSITIVE LOGITS
     München
    0.07
    elerinin
    0.07
     انتخابات
    0.07
    永
    0.07
     vybav
    0.06
    eldorf
    0.06
     Duis
    0.06
     Dost
    0.06
    .m
    0.06
     Europa
    0.06
    Act Density 0.044%

    No Known Activations