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

    Neuronpedia

    INDEX
    Explanations

    Code symbols

    np_max-act · gemini-2.0-flash

    This neuron detects lines where the agent issues an “Action:” directive, i.e. the specification of a tool or command to execute.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    menus
    -0.07
    UY
    -0.06
    ität
    -0.06
    _TIMEOUT
    -0.06
     NEW
    -0.06
    _CENTER
    -0.06
     doch
    -0.06
     eighth
    -0.06
    _EXTERNAL
    -0.06
    anden
    -0.06
    POSITIVE LOGITS
     holiday
    0.06
    .GetFiles
    0.06
     кожного
    0.06
     khẩu
    0.06
     Patrol
    0.06
    (TAG
    0.06
    PING
    0.06
    -football
    0.06
    าก
    0.06
    ávis
    0.06
    Act Density 0.009%

    No Known Activations