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

    Neuronpedia

    INDEX
    Explanations

    database software

    np_max-act · gemini-2.0-flash

    The neuron activates on uppercase configuration identifiers—especially environment‐variable names and other all-caps keys in scripts or YAML.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    chedulers
    -0.07
     outings
    -0.06
     begins
    -0.06
     overlapping
    -0.06
     Tal
    -0.06
    .collect
    -0.06
     }</
    -0.06
     footage
    -0.06
    essential
    -0.06
    _CLIP
    -0.06
    POSITIVE LOGITS
     کمی
    0.07
     zvyš
    0.06
    baar
    0.06
     dụ
    0.06
     hometown
    0.06
    andWhere
    0.06
    kom
    0.06
     강남
    0.06
     sistemi
    0.06
    roat
    0.06
    Act Density 0.018%

    No Known Activations