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

    Neuronpedia

    INDEX
    Explanations

    Describing events/actions

    np_max-act · gemini-2.0-flash

    The neuron is activated by progressive‐form verbs (-ing) and related words that describe a change of state or growing closeness in the relationship.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    άνα
    -0.07
     tình
    -0.07
    .reason
    -0.07
    AGAIN
    -0.06
    '%(
    -0.06
     sø
    -0.06
     Nguyên
    -0.06
    าอ
    -0.06
     AGAIN
    -0.06
    地下
    -0.06
    POSITIVE LOGITS
    actable
    0.08
    oding
    0.08
    encing
    0.07
     Avoid
    0.07
     उ
    0.07
     enlight
    0.06
     declare
    0.06
    ">//
    0.06
    our
    0.06
     आपक
    0.06
    Act Density 0.003%

    No Known Activations