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

    Neuronpedia

    INDEX
    Explanations

    on

    np_max-act · gemini-2.0-flash

    This neuron activates on words from the phrase “get back on track,” i.e. tokens like “get,” “back,” “on,” and “track.”

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     Num
    -0.06
    _HISTORY
    -0.06
    _car
    -0.06
    _agent
    -0.06
     pap
    -0.06
     SPE
    -0.06
     franch
    -0.06
     Zust
    -0.06
    ZD
    -0.06
    <small
    -0.06
    POSITIVE LOGITS
    ịp
    0.07
    την
    0.07
     bilder
    0.06
     hely
    0.06
    entr
    0.06
    (aa
    0.06
    出版
    0.06
    edm
    0.06
    .addRow
    0.06
    touches
    0.06
    Act Density 0.019%

    No Known Activations