Checkpoint Sharding uye Resumable Kudzidzisa Quiz

Mhinduro uye tsananguro zvinoonekwa paunenge uchitora bvunzo.

1. Sei mahombe-emhando yekutarisa akakamurwa kuita shards?
   • Kuita kuti modhi idzidzise nekukurumidza pane zvidiki zvekupinza
   • Kuti uvanze huremu hwemuenzaniso kubva kune vashandisi
   • Nokuti sharding inovandudza kururama
   • Nekuti cheki imwe chete inogona kunge yakakurisa kuti faira rimwechete kana muchina ugone kubata nemazvo
2. Which extra state is typically needed for training resumption beyond model weights?
   • Chekupedzisira chechokwadi chibodzwa
   • Optimizer and scheduler state, step progress, RNG state, and relevant data-loader progress
   • Only a model-weight shard index
   • A screenshot of the training dashboard
3. Ndeipi bhenefiti huru yekudzidzirazve kudzidziswa kwemabasa marefu?
   • Inobvisa kudiwa kweseti yekusimbisa
   • Inovimbisa iyo modhi haimbofi yakapfuura
   • Kupunzika kana kuregererwa hakumanikidze kutangazve kubva pakatanga
   • Inobvisa iyo optimizer
4. In a sharded distributed checkpoint, how can ranks avoid gathering the entire state on one worker?
   • By saving their assigned local tensor shards with checkpoint metadata
   • By each emailing a full model copy
   • By deleting optimizer state before every save
   • By disabling all state serialization
5. Ndeipi basa rinoitwa neindex faira mumasharded checkpoints?
   • Inochengeta password yemushandisi
   • Iyo inomepu zita rega rega reparameter kune shard faira rineyo
   • Iyo inomanikidza mifananidzo mu dataset
   • Inoronga mwero wekudzidza
6. Ndeipi fomati inowanzoshandiswa kune yakachengeteka, ndangariro-mepu yekurodha yemhando shards?
   • MP3
   • CSV
   • kuchengetedza
   • ZIP-chete zvinyorwa
7. A package contains safetensors weight shards and an index, but no optimizer state. What can it support by itself?
   • Loading model weights, without fully restoring optimizer training history
   • Guaranteed exact resumption of every interrupted run
   • Recovering the next batch position automatically
   • Reconstructing every random-number-generator state
8. A sharded checkpoint loader supports a new rank layout. What should the team verify before trusting recovery?
   • That fewer ranks guarantee identical floating-point results
   • That the weight index alone includes every application state
   • That no restart test is necessary
   • That the necessary saved state loads correctly and the resumed training step behaves as expected