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

    Neuronpedia

    INDEX
    Explanations

    5

    np_max-act · gemini-2.0-flash

    The neuron activates on decimal number literals (floating‐point constants) in the code.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    #----------------------------------------------------------------------------
    -0.06
    PLIED
    -0.06
     Graduate
    -0.06
    .Virtual
    -0.06
     Sat
    -0.06
     Dep
    -0.06
     oblig
    -0.06
    .arch
    -0.06
     Sic
    -0.06
    VERIFY
    -0.05
    POSITIVE LOGITS
    壊
    0.07
     leven
    0.07
    세요
    0.07
    螺
    0.06
     loses
    0.06
     náklad
    0.06
     अम
    0.06
     modes
    0.06
    炸
    0.06
    ्सर
    0.06
    Act Density 0.001%

    No Known Activations