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

    Neuronpedia

    INDEX
    Explanations

    code

    np_max-act · gemini-2.0-flash

    The neuron activates on floating-point numeric literals (numbers with a decimal point) in code.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     norm
    -0.07
     indulge
    -0.07
     saison
    -0.07
    Wei
    -0.06
    )')↵
    -0.06
    )")↵
    -0.06
    (pb
    -0.06
     huh
    -0.06
    merican
    -0.06
    Shape
    -0.06
    POSITIVE LOGITS
     neger
    0.06
    -pl
    0.06
    chooser
    0.06
     Yorkers
    0.06
    핑
    0.06
    vable
    0.06
     раб
    0.06
     clr
    0.06
     ExecutionContext
    0.05
     unserer
    0.05
    Act Density 0.050%

    No Known Activations