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

    Neuronpedia

    INDEX
    Explanations

    (JFrame

    np_max-act · gemini-2.0-flash

    The neuron is primarily triggered by floating‐point numeric literals (decimal numbers) in the code.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     pockets
    -0.07
     rud
    -0.06
    ither
    -0.06
     απ
    -0.06
     ICON
    -0.06
    省
    -0.06
     rol
    -0.06
     Ages
    -0.06
     Titles
    -0.06
    Speaking
    -0.06
    POSITIVE LOGITS
    (JFrame
    0.13
    ？”↵↵
    0.07
     liegt
    0.07
     Вар
    0.07
    invitation
    0.07
    leftright
    0.07
    ором
    0.07
     retries
    0.06
     виступ
    0.06
    Recursive
    0.06
    Act Density 0.000%

    No Known Activations