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

    Neuronpedia

    INDEX
    Explanations

    written communication excerpts

    np_max-act · gemini-2.0-flash

    The neuron strongly activates on personal, first-person pronouns and self-references (e.g. “I,” “we,” “my”), marking passages where the author speaks in their own voice.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     LD
    -0.07
    300
    -0.07
    ANGED
    -0.06
    Ticket
    -0.06
    ollen
    -0.06
    警
    -0.06
    PARAM
    -0.06
    ROW
    -0.06
    BX
    -0.06
     такие
    -0.06
    POSITIVE LOGITS
    .Help
    0.07
     hern
    0.07
     adherence
    0.07
     Sitting
    0.06
     stripped
    0.06
     ceil
    0.06
     Kurdistan
    0.06
     degli
    0.06
    套
    0.06
     hintText
    0.06
    Act Density 0.151%

    No Known Activations