Quiz ci model yu jaxaso ak yu néew

Sooy def quiz bi, tontu yi ak leeral yi dañuy feeñ.

1. Ci Mixtral 8x7B, ñaata kàngam ñooy liggéey ci token bu nekk?
   • Ñaari kàngam
   • Juróom ñatti kàngam yépp
   • Benn eksper
   • Nimero bu bari saa yu nekk
2. Ban cër mooy dogal ban kàngam lañuy yónnee token bi?
   • Tokenisatër bi
   • Routeur bi (réseau gating)
   • Couche d'attention
   • Tablo biy dugal
3. Doonte Mixtral 8x7B amna lu tollu ci 47B ci parametre yi, lu tollu ci ñaata lañu ci token bu nekk?
   • Lu tollu ci 47 milyaar
   • Lu tollu ci 7 milyaar
   • Lu tollu ci 13 milyaar
   • Lu tollu ci 70 milyaar
4. What must a Mixtral serving setup account for even when only two experts run per token?
   • Only the active parameters ever need to be stored anywhere
   • All experts must always fit in GPU memory with no offloading option
   • Unused experts can be deleted without changing the checkpoint
   • The full expert weights must be available to the system, with residency or offloading depending on deployment
5. Ci ban lisence la Mistral AI génne Mixtral, tax ñu nangu ko?
   • Apache 2.0
   • Boroom, xët bu tëju kese
   • GPL v3
   • Creative Commons du njaay
6. What training steps does Mistral report for the Mixtral 8x7B Instruct variant?
   • Dina dindi bépp soxla ci done tàggat yaram
   • Supervised fine-tuning and direct preference optimization (DPO)
   • Dafay dindi soxla GPUs lëmm
   • Dafay garanti ni model yi duñu musa def benn njuumte
7. How many expert networks are available in each Mixtral 8x7B sparse mixture-of-experts layer?
   • Ñaari kàngam
   • Seven experts
   • Eight experts
   • Twelve experts
8. Which languages does Mistral list for Mixtral 8x7B?
   • Yes; all unused expert weights are deleted
   • Yes; it means the model is dense with 12.9B total parameters
   • No; it means every parameter is active for every token
   • English, French, Italian, German, and Spanish