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

    Neuronpedia

    INDEX
    Explanations

    Blog post excerpts

    np_max-act · gemini-2.0-flash

    This neuron primarily detects section or subsection headings (i.e. title‐case phrases set off as headings).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    (loss
    -0.07
     ورد
    -0.07
    ,column
    -0.07
     seating
    -0.07
    Popular
    -0.06
    .Signal
    -0.06
     parametros
    -0.06
    织
    -0.06
    study
    -0.06
    payments
    -0.06
    POSITIVE LOGITS
    ayette
    0.06
    ा↵↵
    0.06
     Ole
    0.06
     Type
    0.06
    NET
    0.06
    %m
    0.06
     Arena
    0.06
     yaşan
    0.06
     Rhino
    0.06
     MON
    0.06
    Act Density 0.036%

    No Known Activations