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

    Neuronpedia

    INDEX
    Explanations

    email, especially spam

    np_max-act · gemini-2.0-flash

    The neuron detects instructions telling the reader to check their spam or junk email folder.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     incom
    -0.07
     Validation
    -0.06
    Axes
    -0.06
     neck
    -0.06
    lectual
    -0.06
     IID
    -0.06
    %↵
    -0.06
    組
    -0.06
     Intellectual
    -0.06
     fulfilled
    -0.06
    POSITIVE LOGITS
    евой
    0.07
    	back
    0.07
    .RestController
    0.07
    .Extensions
    0.06
     прибор
    0.06
     underground
    0.06
     DEV
    0.06
     spoj
    0.06
     predictor
    0.06
    ротив
    0.06
    Act Density 0.020%

    No Known Activations