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

    Neuronpedia

    INDEX
    Explanations

    posture

    np_max-act · gemini-2.0-flash

    This neuron responds to mentions of “posture,” activating strongly on the word and its subtoken parts.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     یون
    -0.07
     rapide
    -0.07
    (manager
    -0.07
    racuse
    -0.07
    thritis
    -0.06
    _wait
    -0.06
    SingleNode
    -0.06
    =value
    -0.06
     bombs
    -0.06
    Cleanup
    -0.06
    POSITIVE LOGITS
     posture
    0.09
    	UI
    0.08
    /Input
    0.07
     struct
    0.06
    ��
    0.06
    �
    0.06
    중
    0.06
    struct
    0.06
     sudden
    0.06
     Ruf
    0.06
    Act Density 0.005%

    No Known Activations