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

    Neuronpedia

    INDEX
    Explanations

    seed

    np_max-act · gemini-2.0-flash

    This neuron detects mentions of playoff or tournament seedings and bracket positions (e.g., team “seed,” “seedings,” and related ranking numbers).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    910
    -0.07
     toys
    -0.07
    afi
    -0.06
    umbled
    -0.06
     begging
    -0.06
    afari
    -0.06
    显
    -0.06
    رق
    -0.06
    ضوع
    -0.06
     огра
    -0.06
    POSITIVE LOGITS
     getPassword
    0.07
     (~(
    0.06
     нашей
    0.06
    ão
    0.06
    류
    0.06
     Enhanced
    0.06
     GLES
    0.06
     "$
    0.06
    .AD
    0.06
     letterSpacing
    0.06
    Act Density 0.005%

    No Known Activations