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

    Neuronpedia

    INDEX
    Explanations

    Code brackets

    np_max-act · gemini-2.0-flash

    This neuron activates on programming‐language syntax and identifiers (i.e. source‐code tokens) rather than natural‐language text.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    lif
    -0.06
    Downloader
    -0.06
    wash
    -0.06
    pres
    -0.06
     flavours
    -0.06
    vg
    -0.06
     rio
    -0.06
     мощ
    -0.06
    eres
    -0.05
    bios
    -0.05
    POSITIVE LOGITS
     Warren
    0.08
     AttributeSet
    0.07
     sum
    0.07
    نا
    0.07
     ему
    0.07
     添加
    0.07
     بد
    0.07
     taxable
    0.07
     میل
    0.07
    .resolve
    0.06
    Act Density 0.057%

    No Known Activations