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

    Neuronpedia

    INDEX
    Explanations

    code and technical text

    np_max-act · gemini-2.0-flash

    This neuron activates on tokens that are parts of file‐system paths (e.g. Windows drive letters, backslashes, directory and filename segments).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    لاة
    -0.07
     choking
    -0.06
    .Features
    -0.06
     Lama
    -0.06
    Before
    -0.06
    移动
    -0.06
    Non
    -0.06
    dictions
    -0.06
    multiple
    -0.06
     meas
    -0.06
    POSITIVE LOGITS
    �
    0.06
    상의
    0.06
    暮
    0.06
     oč
    0.06
    .soft
    0.06
    直
    0.06
    ','#
    0.06
     подт
    0.05
    _periods
    0.05
    Expose
    0.05
    Act Density 0.991%

    No Known Activations