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

    Neuronpedia

    INDEX
    Explanations

    varied text excerpts

    np_max-act · gemini-2.0-flash

    This neuron activates on French-language words and phrasing, effectively detecting segments written in French.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     pollution
    -0.08
     greens
    -0.07
     screen
    -0.06
     Screen
    -0.06
    archy
    -0.06
     visions
    -0.06
     rims
    -0.06
    ジェ
    -0.06
     motivation
    -0.06
    附
    -0.06
    POSITIVE LOGITS
     endings
    0.07
    ='"+
    0.07
    (groups
    0.07
     eql
    0.07
     stk
    0.06
    enido
    0.06
     груп
    0.06
    ?↵↵↵↵
    0.06
     gele
    0.06
    <pcl
    0.06
    Act Density 0.130%

    No Known Activations