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

    Neuronpedia

    INDEX
    Explanations

    }

    np_max-act · gemini-2.0-flash

    This neuron detects in‐text figure (and similar) reference markers (e.g. the backslash or bracketed “Figure #” citations).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     Randall
    -0.07
     FileOutputStream
    -0.06
     Joel
    -0.06
    eru
    -0.06
     --------------------
    -0.06
    .es
    -0.06
    čen
    -0.06
    еч
    -0.06
    	C
    -0.06
    оки
    -0.06
    POSITIVE LOGITS
     blessing
    0.07
     phenomenon
    0.07
    .DataVisualization
    0.07
    шая
    0.07
     networking
    0.06
     jar
    0.06
    이었
    0.06
     tendency
    0.06
     lack
    0.06
     rearr
    0.06
    Act Density 0.003%

    No Known Activations