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

    Neuronpedia

    INDEX
    Explanations

    personal reflections

    np_max-act · gemini-2.0-flash

    This neuron activates on first-person self-referential language (pronouns like “I,” “me,” “my” and related introspective expressions).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     سایت
    -0.07
    houses
    -0.07
    گار
    -0.06
     Fear
    -0.06
     codigo
    -0.06
    Tuple
    -0.06
     ideals
    -0.06
    esi
    -0.06
    stamp
    -0.06
    EVENT
    -0.06
    POSITIVE LOGITS
     //--
    0.07
    ंब
    0.07
     Ops
    0.07
    UNUSED
    0.07
    Connector
    0.06
     taskId
    0.06
     srpna
    0.06
    StartupScript
    0.06
    (ns
    0.06
     رض
    0.06
    Act Density 0.047%

    No Known Activations