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

    Neuronpedia

    INDEX
    Explanations

    code

    np_max-act · gemini-2.0-flash

    This neuron detects the “Cause:” label (i.e. the token indicating the start of a cause field).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    스트
    -0.07
    ========
    -0.07
    Connect
    -0.07
    .lst
    -0.06
    ěž
    -0.06
    	lines
    -0.06
    write
    -0.06
     rape
    -0.06
     unavoidable
    -0.06
    press
    -0.06
    POSITIVE LOGITS
    ��
    0.07
     UIButton
    0.07
    HasColumnName
    0.07
     kak
    0.06
    obic
    0.06
     Mojo
    0.06
     Joe
    0.06
    ิร
    0.06
     �
    0.06
    ून
    0.06
    Act Density 0.011%

    No Known Activations