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

    Neuronpedia

    INDEX
    Explanations

    arrays and lists

    np_max-act · gemini-2.0-flash

    The neuron fires on programming-style tokens and constructs (e.g. variable names, keywords, API calls, punctuation) and so essentially detects segments of source code.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    َف
    -0.07
    (arg
    -0.07
     wasting
    -0.06
    (fake
    -0.06
    (signal
    -0.06
     "\↵
    -0.06
    [tmp
    -0.06
    .RequestParam
    -0.06
     astronomy
    -0.06
     smokers
    -0.06
    POSITIVE LOGITS
    été
    0.07
     semaine
    0.07
    Alignment
    0.06
     كامل
    0.06
    _trace
    0.06
    -ignore
    0.06
    :new
    0.06
     فرمان
    0.06
     QtCore
    0.06
     sorts
    0.06
    Act Density 0.044%

    No Known Activations