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

    Neuronpedia

    INDEX
    Explanations

    .

    np_max-act · gemini-2.0-flash

    This neuron selectively activates on Arabic-script tokens.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     pe
    -0.07
    .onSubmit
    -0.07
     mou
    -0.06
    (timer
    -0.06
     xa
    -0.06
    itre
    -0.06
     kent
    -0.06
    .onDestroy
    -0.06
     z
    -0.06
    OUGH
    -0.06
    POSITIVE LOGITS
    .enum
    0.06
    	job
    0.06
     �
    0.06
     Tutor
    0.06
    /Object
    0.06
    enga
    0.06
     đồ
    0.06
     شیر
    0.05
     telesc
    0.05
    depend
    0.05
    Act Density 0.028%

    No Known Activations