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

    Neuronpedia

    INDEX
    Explanations

    Game command sequences

    np_max-act · gemini-2.0-flash

    The neuron specifically detects the model’s internal protocol/control tokens (e.g. metadata markers like <|eot_id|>, header delimiters, and other non‐content structural tags).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     arranging
    -0.07
     Greatest
    -0.06
    igor
    -0.06
     importantes
    -0.06
     sending
    -0.06
    ########################################################
    -0.06
     accelerated
    -0.06
    mitter
    -0.06
     drinking
    -0.06
    -best
    -0.06
    POSITIVE LOGITS
    (edit
    0.08
    moz
    0.07
    řít
    0.07
    .orig
    0.07
    ليات
    0.06
     moss
    0.06
    azione
    0.06
    ACKET
    0.06
     느�
    0.06
     České
    0.06
    Act Density 0.028%

    No Known Activations