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

    Neuronpedia

    INDEX
    Explanations

    Actions and instructions

    np_max-act · gemini-2.0-flash

    The neuron strongly responds to the special end‐of‐turn/text token (e.g. “<|eot_id|>”).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    レビ
    -0.06
     те
    -0.06
     derec
    -0.06
    ült
    -0.06
     pre
    -0.06
     verte
    -0.06
    imary
    -0.06
    uzione
    -0.06
    _exclude
    -0.06
     Nas
    -0.06
    POSITIVE LOGITS
     slot
    0.07
     ok
    0.07
    lj
    0.06
    Datum
    0.06
    .mapbox
    0.06
    ۱۹۸
    0.06
     dams
    0.06
     Pil
    0.06
     sms
    0.06
    $/
    0.06
    Act Density 0.038%

    No Known Activations