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

    Neuronpedia

    INDEX
    Explanations

    code

    np_max-act · gemini-2.0-flash

    Explanation of neuron 4 behavior: the main thing this neuron does is find numeric tokens (numbers and version/float literals) in code and text.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     처
    -0.07
    UMP
    -0.06
    ований
    -0.06
     damage
    -0.06
    *))
    -0.06
    Unlock
    -0.06
     pulses
    -0.06
     unders
    -0.05
     neutrality
    -0.05
     gamble
    -0.05
    POSITIVE LOGITS
    :")↵
    0.07
     pageTitle
    0.07
    .normalized
    0.07
    .↵↵↵↵↵↵↵↵↵↵↵↵
    0.07
    页面存档备份
    0.06
     Vladim
    0.06
    .getActive
    0.06
     sayılı
    0.06
    ٹ
    0.06
    ->_
    0.06
    Act Density 0.037%

    No Known Activations