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

    Neuronpedia

    INDEX
    Explanations

    in

    np_max-act · gemini-2.0-flash

    The neuron activates specifically on the literal “\[” bracket token that marks the start of the user-provided answer placeholder.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    cerr
    -0.06
    边
    -0.06
     Carol
    -0.06
    .rar
    -0.06
     sideline
    -0.06
     Toni
    -0.06
    .AspNetCore
    -0.06
    %"↵
    -0.06
    .cn
    -0.06
     Poe
    -0.06
    POSITIVE LOGITS
     genre
    0.07
    /icons
    0.07
    .wikipedia
    0.07
     розрах
    0.07
     concluding
    0.06
     aqu
    0.06
    _language
    0.06
     philosophical
    0.06
    .Align
    0.06
     millions
    0.06
    Act Density 0.001%

    No Known Activations