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

    Neuronpedia

    INDEX
    Explanations

    project initiation/completion

    np_max-act · gemini-2.0-flash

    This neuron selectively activates on formal “event” verbs that announce or enact official actions (e.g. enacted, approved, launched, announced, initiated).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    (embed
    -0.07
    Maintenance
    -0.06
    _supplier
    -0.06
    aled
    -0.06
     commanders
    -0.06
    Viewer
    -0.06
     Knot
    -0.06
     Nov
    -0.06
     rationale
    -0.06
    /Public
    -0.06
    POSITIVE LOGITS
    Vien
    0.07
    .twig
    0.07
    」「
    0.07
     zb
    0.07
    IGHT
    0.06
     VERY
    0.06
    ');?>↵
    0.06
    appName
    0.06
    ')?>
    0.06
    ।
    0.06
    Act Density 0.032%

    No Known Activations