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

    Neuronpedia

    INDEX
    Explanations

    protests and police

    np_max-act · gemini-2.0-flash

    This neuron detects mentions of police or law‐enforcement actions (e.g. “police,” “held,” “kettled,” “arrest,” etc.).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     Gat
    -0.06
     Fak
    -0.06
     grid
    -0.06
     dias
    -0.06
     Dick
    -0.06
     dress
    -0.06
    rip
    -0.06
    il
    -0.06
    yaw
    -0.06
     beginner
    -0.06
    POSITIVE LOGITS
     บาท
    0.07
     Controlled
    0.07
     newY
    0.06
     t�
    0.06
    "go
    0.06
     мик
    0.06
     tbody
    0.06
     små
    0.06
    .MONTH
    0.06
    ETER
    0.06
    Act Density 0.004%

    No Known Activations