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

    Neuronpedia

    INDEX
    Explanations

    forum posts

    np_max-act · gemini-2.0-flash

    The neuron activates on date/time stamps (numbers and separators in post headers), i.e. fragments of timestamps.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     uninitialized
    -0.07
     explosion
    -0.07
    <Action
    -0.06
     ruined
    -0.06
    fold
    -0.06
     完
    -0.06
    ョ
    -0.06
     loan
    -0.06
     inference
    -0.06
    ाहन
    -0.06
    POSITIVE LOGITS
    orrar
    0.06
     kèo
    0.06
    eters
    0.06
     local
    0.06
    etic
    0.06
     Henri
    0.06
    _MPI
    0.06
     свящ
    0.06
    posed
    0.06
    etics
    0.06
    Act Density 0.005%

    No Known Activations