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

    Neuronpedia

    INDEX
    Explanations

    crew

    np_max-act · gemini-2.0-flash

    The neuron fires on tokens naming social parties or relationships (e.g. crew, strains) – in other words, it detects words referring to people‐group entities or their interpersonal dynamics.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     McCorm
    -0.06
    617
    -0.06
    .same
    -0.06
     Νο
    -0.06
    hits
    -0.06
     fis
    -0.06
     알고
    -0.06
    lsruhe
    -0.06
     onTap
    -0.06
     knot
    -0.06
    POSITIVE LOGITS
     Johnny
    0.07
     guarda
    0.07
    .get
    0.07
     dying
    0.06
    ByID
    0.06
    .getToken
    0.06
    (point
    0.06
     Config
    0.06
    )?↵↵
    0.06
    	Double
    0.06
    Act Density 0.000%

    No Known Activations