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

    Neuronpedia

    INDEX
    Explanations

    Here

    np_max-act · gemini-2.0-flash

    The neuron activates on direct second‐person commands or invitations (imperative phrases like “come here,” “sit down,” or “let me…”).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    ceans
    -0.07
     descended
    -0.06
    _syn
    -0.06
     sights
    -0.06
    .lua
    -0.06
     神马
    -0.06
     نیست
    -0.06
     mounted
    -0.06
     oversight
    -0.06
     tackled
    -0.06
    POSITIVE LOGITS
    Disallow
    0.07
     "'.$
    0.07
    imonial
    0.07
    OperationContract
    0.07
    -effective
    0.06
     socioeconomic
    0.06
     ).
    0.06
    ».↵
    0.06
    mental
    0.06
    FromFile
    0.06
    Act Density 0.010%

    No Known Activations