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

    Neuronpedia

    INDEX
    Explanations

    quantities and scales

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ந
    0.44
     schoolchildren
    0.43
    aparikkh
    0.42
    ŠE
    0.42
     schul
    0.42
    ẫ
    0.42
     bewildered
    0.41
    Türk
    0.40
    În
    0.40
     евре
    0.40
    POSITIVE LOGITS
     τόσο
    0.46
     begitu
    0.44
    prices
    0.42
     prices
    0.41
     then
    0.40
    years
    0.40
     Voilà
    0.39
     намного
    0.39
     nemmeno
    0.39
     tanta
    0.39
    Act Density 0.029%

    No Known Activations