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

    Neuronpedia

    INDEX
    Explanations

    Non-English

    np_max-act · gemini-2.0-flash

    The neuron activates on contiguous Arabic-script text, effectively detecting segments written in Arabic.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     tour
    -0.08
     parler
    -0.07
     separation
    -0.06
    яття
    -0.06
    erten
    -0.06
     Igor
    -0.06
    是我
    -0.06
     dice
    -0.06
     América
    -0.06
     ['
    -0.06
    POSITIVE LOGITS
    CppTypeDefinitionSizes
    0.08
    agog
    0.07
     без
    0.07
    _Zero
    0.07
    serde
    0.06
    reamble
    0.06
    'LBL
    0.06
    âte
    0.06
    .tbl
    0.06
    	End
    0.06
    Act Density 0.038%

    No Known Activations