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

    Neuronpedia

    INDEX
    Explanations

    comparing two different things

    np_max-act · gemini-2.0-flash

    The neuron detects the special placeholder tokens used to label characters (e.g. “NAME_2,” “NAME_3,” etc.).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     mi
    -0.08
     desarroll
    -0.07
    .Return
    -0.07
    .Large
    -0.07
    -sided
    -0.07
     Manager
    -0.07
     polynomial
    -0.06
    YS
    -0.06
     раск
    -0.06
     PRIV
    -0.06
    POSITIVE LOGITS
    iena
    0.06
     граду
    0.06
    ,filename
    0.06
    (display
    0.06
     hardship
    0.06
    (*)(
    0.06
    esModule
    0.06
    .INTEGER
    0.06
     числе
    0.06
    (&_
    0.05
    Act Density 0.014%

    No Known Activations