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

    Neuronpedia

    INDEX
    Explanations

    success and failure outcomes

    np_max-act · gemini-2.0-flash

    The neuron is primarily activated by numeric tokens (especially decimal numbers) and so is detecting numerical/statistical values in the text.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    wk
    -0.07
     Tic
    -0.07
    짜
    -0.06
    Cycle
    -0.06
    ﾍ
    -0.06
    .Vert
    -0.06
    ())),
    -0.06
    .live
    -0.06
    _files
    -0.06
    -wsj
    -0.06
    POSITIVE LOGITS
     victorious
    0.07
     advertising
    0.07
    enerator
    0.07
    (columns
    0.06
     Joe
    0.06
     Lagos
    0.06
     ní
    0.06
    .RemoveAll
    0.06
    igrate
    0.06
     Ministry
    0.06
    Act Density 0.005%

    No Known Activations