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

    Neuronpedia

    INDEX
    Explanations

    past events

    np_max-act · gemini-2.0-flash

    This neuron detects expressions referring to a previous occasion, especially the phrase “the last time.”

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    afc
    -0.07
    oft
    -0.07
    rain
    -0.06
    .shapes
    -0.06
    aft
    -0.06
     applicant
    -0.06
    fdf
    -0.06
    Nh
    -0.06
     make
    -0.06
     Pink
    -0.06
    POSITIVE LOGITS
     уча
    0.07
    ').'</
    0.06
    ?????
    0.06
    ()))↵↵
    0.06
     "'"
    0.06
    bart
    0.06
    ála
    0.06
    #↵↵
    0.06
    pit
    0.06
    ("^
    0.06
    Act Density 0.028%

    No Known Activations