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

    Neuronpedia

    INDEX
    Explanations

    quotation marks

    np_max-act · gemini-2.0-flash

    This neuron detects the opening quotation mark that begins a YAML front-matter string (e.g. the quote before a title value).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    .spin
    -0.07
     Restoration
    -0.07
     Near
    -0.07
     TOR
    -0.06
    arehouse
    -0.06
    оту
    -0.06
     occurrence
    -0.06
     màn
    -0.06
    二人
    -0.06
     Gods
    -0.06
    POSITIVE LOGITS
     Nack
    0.07
     adap
    0.07
    _attrib
    0.06
    \Message
    0.06
     discre
    0.06
     vốn
    0.06
    .Payload
    0.06
     inconsistencies
    0.06
     oluşan
    0.06
     соедин
    0.06
    Act Density 0.000%

    No Known Activations