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

    Neuronpedia

    INDEX
    Explanations

    ump

    np_max-act · gemini-2.0-flash

    This neuron activates on the metadata lines inserted by the class‐dump tool—i.e., the occurrence of “dump” in the “Generated by class-dump” comments.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    りの
    -0.08
     chez
    -0.07
     fik
    -0.07
     afflict
    -0.07
     TIFF
    -0.07
    ampion
    -0.07
    gia
    -0.07
     pension
    -0.07
     cleric
    -0.07
     rend
    -0.07
    POSITIVE LOGITS
    dataTable
    0.06
    _LVL
    0.06
     "</
    0.06
    kyně
    0.06
    .getClassName
    0.06
    claim
    0.06
    _OscInitStruct
    0.06
     viewController
    0.06
     config
    0.05
    (stderr
    0.05
    Act Density 0.000%

    No Known Activations