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

    Neuronpedia

    INDEX
    Explanations

    brackets

    np_max-act · gemini-2.0-flash

    This neuron activates on the JSON structural punctuation (braces and quotation marks) used when formatting output as JSON.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    ave
    -0.07
     };
    -0.07
     Ruiz
    -0.06
    -exec
    -0.06
     ков
    -0.06
     Schultz
    -0.06
     healer
    -0.06
    о�
    -0.06
    drv
    -0.06
    □
    -0.06
    POSITIVE LOGITS
     또
    0.07
    ีความ
    0.06
    ुत
    0.06
    (t
    0.06
     tempfile
    0.06
    ERR
    0.06
    茂
    0.06
    ?>">↵
    0.06
    ротив
    0.06
    .Product
    0.06
    Act Density 0.017%

    No Known Activations