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

    Neuronpedia

    INDEX
    Explanations

    around

    np_max-act · gemini-2.0-flash

    This neuron fires on framing phrases that center or build something “around” a concept (e.g. “built around,” “structured around,” “centered around”).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    =json
    -0.07
     Flask
    -0.07
    Usage
    -0.07
     inspect
    -0.07
     bring
    -0.07
     nhưng
    -0.07
     bringing
    -0.07
     @"↵
    -0.07
    .Http
    -0.06
    isch
    -0.06
    POSITIVE LOGITS
     principales
    0.06
     ร
    0.06
    character
    0.06
    dee
    0.06
     McDonald
    0.06
     IData
    0.06
    tol
    0.06
     nhánh
    0.06
    UpDown
    0.06
    lashes
    0.06
    Act Density 0.039%

    No Known Activations