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

    Neuronpedia

    INDEX
    Explanations

    illegitimate keys

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     î
    0.50
     ነው።
    0.49
    ên
    0.49
    重要な
    0.49
     önemli
    0.48
     acompaña
    0.48
     keç
    0.47
    醗
    0.47
     leaking
    0.47
     മേ
    0.46
    POSITIVE LOGITS
    l
    0.50
     Attractions
    0.47
    BERT
    0.45
    awatir
    0.45
    StringSet
    0.44
    DBOutput
    0.44
    salaryfrom
    0.43
    antra
    0.43
    Charter
    0.43
    Output
    0.42
    Act Density 0.000%

    No Known Activations