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

    Neuronpedia

    INDEX
    Explanations

    in

    np_max-act · gemini-2.0-flash

    This neuron detects tokens representing the verb “arrive” (and its morphological variants) indicating arrival.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     governors
    -0.07
     chocolate
    -0.07
    metal
    -0.07
    .Sum
    -0.07
     TORT
    -0.07
     hostility
    -0.07
    -makers
    -0.06
    ("'",
    -0.06
    aya
    -0.06
    -Feb
    -0.06
    POSITIVE LOGITS
     }
↵
↵
↵
    0.07
    ret
    0.07
    vide
    0.06
     customize
    0.06
    )
↵
↵
↵
    0.06
    })↵↵↵
    0.06
    .Pages
    0.06
     //
↵
↵
    0.06
    /'↵
    0.06
    .');↵
    0.06
    Act Density 0.041%

    No Known Activations