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

    Neuronpedia

    INDEX
    Explanations

    academic/technical writing

    np_max-act · gemini-2.0-flash

    The neuron detects instructional or “how-to” phrasing that offers guidance or explanations.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    _shutdown
    -0.07
     Beverly
    -0.07
    -0.07
     isempty
    -0.07
    .servers
    -0.07
    onz
    -0.06
    ogi
    -0.06
    在地
    -0.06
    ...";↵
    -0.06
    .Created
    -0.06
    POSITIVE LOGITS
    erule
    0.06
    _DEF
    0.06
    abilidad
    0.06
     Chase
    0.06
    arton
    0.06
     mistress
    0.06
    struct
    0.06
     zásob
    0.05
    aysia
    0.05
     beasts
    0.05
    Act Density 0.186%

    No Known Activations