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

    Neuronpedia

    INDEX
    Explanations

    eating

    np_max-act · gemini-2.0-flash

    This neuron detects French-language words (i.e. it activates on French tokens).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     chicks
    -0.07
    eson
    -0.07
    dd
    -0.07
    .De
    -0.07
    ческих
    -0.07
    .Pop
    -0.07
     Blob
    -0.07
     Care
    -0.06
     Dodd
    -0.06
    Ta
    -0.06
    POSITIVE LOGITS
     sebeb
    0.07
    ść
    0.07
     ammunition
    0.07
     vaping
    0.06
     challeng
    0.06
    .setObjectName
    0.06
    	freopen
    0.06
    UseProgram
    0.06
     republik
    0.06
    μβ
    0.06
    Act Density 0.045%

    No Known Activations