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

    Neuronpedia

    INDEX
    Explanations

    thoughts

    np_max-act · gemini-2.0-flash

    The neuron responds to discourse markers and filler phrases that express commentary, speculation, or anticipation (e.g. “This is going to,” “hmm,” “oh well,” “maybe it’s”).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    .getStatus
    -0.07
     “
    -0.06
     ولك
    -0.06
     определен
    -0.06
    rer
    -0.06
    肯定
    -0.06
     rovněž
    -0.06
    soever
    -0.06
    اهد
    -0.06
    agree
    -0.06
    POSITIVE LOGITS
    umi
    0.06
    ního
    0.06
     kapas
    0.06
     aşam
    0.06
     створ
    0.06
     elimin
    0.06
     MW
    0.06
    zenie
    0.06
    106
    0.06
    (ctx
    0.06
    Act Density 0.065%

    No Known Activations