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

    Neuronpedia

    INDEX
    Explanations

    sneaking

    np_max-act · gemini-2.0-flash

    The neuron detects structural and formatting markers (special tokens indicating headers, section boundaries, or control delimiters).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    Spe
    -0.07
    Inverse
    -0.06
    InRange
    -0.06
    _counters
    -0.06
     rejoice
    -0.06
    Preparing
    -0.06
    Marcus
    -0.06
    acin
    -0.06
    -independent
    -0.06
    東京
    -0.06
    POSITIVE LOGITS
    _COMPANY
    0.07
     �
    0.07
    ..........
    0.07
     yana
    0.07
    _SAVE
    0.07
     sahibi
    0.06
     －
    0.06
     ⓘ
    0.06
     yani
    0.06
     enquiries
    0.06
    Act Density 0.023%

    No Known Activations