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

    Neuronpedia

    INDEX
    Explanations

    emergency

    np_max-act · gemini-2.0-flash

    This neuron detects mentions of emergency or emergency‐related situations.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     [=
    -0.08
     silicon
    -0.06
     disc
    -0.06
     시스템
    -0.06
    (course
    -0.06
     shader
    -0.06
    нил
    -0.06
     territorial
    -0.06
     ba
    -0.06
    izzo
    -0.06
    POSITIVE LOGITS
     urgent
    0.10
     Emergency
    0.07
    Emergency
    0.07
     vital
    0.07
    Emer
    0.07
    γγ
    0.07
     emergency
    0.07
    )?↵↵
    0.06
    +s
    0.06
     important
    0.06
    Act Density 0.008%

    No Known Activations