A recurrent model trained decentralized, with no blocking and almost no quality loss
About SN64 Chutes
We trained a pure recurrent model across distributed GPUs with fully non-blocking synchronization, and it landed within 0.6% of the centralized baseline at matched steps. As far as we know, that is the first time decentralized non-blocking training has held up on a recurrent architecture. The tradeoff we set out to break Training a model across many GPUs in different places forces a familiar choice. Either the GPUs stop and wait to synchronize with each other, which is slow and expensive, or
Read the article at SN64 Chutes blog ↗
SN64 Chutes at a glance
SN64, Chutes (α token ش), is one of the independent markets on the Bittensor network, each with its own α token and its own price. Its α currently trades at 0.066730 τ (≈ $17.74), against 198,743 τ of pool liquidity. On-chain, the best-delegated validator on it yields about 33.84% APR (simple) and 40.28% APY (compounded), emissions add 0.008768 τ to the pool each block, epochs run every 360 blocks. It was registered at block #4,531,295.
More from SN64 Chutes
Taosis indexes announcements, articles, posts and releases about Bittensor subnets from their own channels and keeps a record of each. The text above is the source’s; the figures are read from the chain by Taosis.