specific method · filed under optimization
Cross-replica model-weight hash consistency checks
Periodically hashes weights across replicas to verify that they remain bit-identical.
Also called Cross-replica Hash Checks.
- source
- 1
- model
- 1
- lab adopt it
- 1
- strongest
- used
How sources treat it
One count per evidence span, weakest treatment to strongest.
used 1
Documented in
Evidence
1 span quoted from the sources, strongest treatment first.
we periodically hash model weights across replicas to assert that they remain bit-identical.
usedevaluation onlyin Model FactoryPoolside
Filed alongside
Other methods under optimization :: training stability.
Weight paddingActivation or logit clippingAnti-hallucination trainingBehavioral regularizationElevated constant-learning-rate training stability stress testExponential moving average of checkpointsGradient clippingLearning-rate elevation for training stability stress testingLoss masking for excessively stale tokensOff-policy sample filteringPer-token regularization for off-policy RLSafeguards against training drift and reward hackingSoft droppingTraining stability stress testingWeight clippingWeight decay coupled to learning-rate squared