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

    Neuronpedia

    INDEX
    Explanations

    Knowing or not knowing

    np_max-act · gemini-2.0-flash

    This neuron responds to words describing personal insight or the discovery of previously unknown qualities (e.g. “never,” “knew,” “hidden,” “creativity,” “learned”).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     empower
    -0.07
     undefeated
    -0.07
     charm
    -0.06
    _zip
    -0.06
    _MOUNT
    -0.06
    flash
    -0.06
     counties
    -0.06
    洪
    -0.06
    239
    -0.06
     Overwatch
    -0.06
    POSITIVE LOGITS
    storage
    0.07
     Bates
    0.07
    light
    0.06
    .Float
    0.06
     bleach
    0.06
    gross
    0.06
    .permissions
    0.06
     Enforcement
    0.06
     gross
    0.06
    nex
    0.06
    Act Density 0.106%

    No Known Activations