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

    Neuronpedia

    INDEX
    Explanations

    copying

    np_max-act · gemini-2.0-flash

    This neuron detects technical references to image‐forming hardware (e.g. copying machines, scanners, printers) in patent text.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    .ORDER
    -0.06
    ,No
    -0.06
    ّ
    -0.06
     Sad
    -0.05
     InkWell
    -0.05
    .HTML
    -0.05
     tiế
    -0.05
    foon
    -0.05
    ierung
    -0.05
    ALLOW
    -0.05
    POSITIVE LOGITS
     thrott
    0.07
     mechanic
    0.07
     tech
    0.07
     Wired
    0.07
    threshold
    0.07
     republic
    0.07
     São
    0.06
     tempt
    0.06
     Username
    0.06
    mine
    0.06
    Act Density 0.005%

    No Known Activations