Mining for Bounty is open. Hypertrain is in preview.

HOW TO MINE
/HYPERTRAIN · CHALLENGE ID hypertrain

Hypertrain. One model, many miners.

Subnet 100 miners train one shared model together with decentralized training, syncing once per round. The figures below are read from the challenge's public API. No reward has matured on these runs, so nothing has been paid yet. The v2 runs use a test ledger.

RUN AT A GLANCE
FINAL LOSS—No round synced.
OUTER ROUND0/ 1The run stopped after 0 of 1 rounds.
CLUSTERS1took partClusters that took part in this run. 1 node, stopped with the run.
AVG THROUGHPUT0tok/sTokens per second over the run, summed over the clusters that took part.
BANDWIDTH SAVED—vs DDPLess data crossed the network than DDP would have sent for the same rounds.
TOKENS TRAINED (FINAL)0/ 0— of the token target when the run stopped.
RUN #0e232dba54072e314930f6f12686a0466a77528ebc384d933a62d6ec6a34ba07Stagelive dataModelOpenDecision od-large (354M)StatusstoppedOuterNesterov · lr 0.70 · μ 0.90SAMPLED 11-22
TRAINING CURVE · GLOBAL LOSS

Is the shared model getting better?

Loss is how wrong the model is on held-out text. Every dot is one outer round; lower is better.
GLOBAL LOSS · ONE DOT PER OUTER ROUND
FINAL LOSS—
SINCE ROUND 1—
LOWEST—
HIGHEST AFTER R1—
STARTED AT—
SYNCED ROUNDS0 of 1
NOT TRAININGFigures as served by the challenge's public API.
TOKENS / S
TOKENS TRAINED

No round has synced yet, so there is no curve to draw.

Hover the curve or focus it and press the arrow keys to read a round.
THE GROUPS · CLUSTERS

Who trained this run?

A cluster is a group of machines that trains together inside one location. Each one does its own local steps between syncs.
CLUSTERS · MACHINES THAT TRAIN TOGETHER
Each square is one GPU. Each block is one node.FINISHED
unassignedunassignedOFFLINE
unknownRTX 5090
NODES · GPUS
1 · 8
THROUGHPUT
—
INNER STEPS THIS ROUND0 / 64
LAST SYNC
round 010-17 00:20 UTC
TOKEN SHARE0.0%
0 tokens
THE ROSTER · PARTICIPANTS AND ROUNDS

Every miner, every round.

Tokens are the work a miner contributed. Share is its part of all tokens trained so far.
PARTICIPANTS · EVERY MINER IN THE RUN
#MINERCLUSTERGPUSSTATUSROUNDSTOKENSTOK/SSHARE
15Do8HW...3Keh5Do8…3Kehunassigned8 x RTX 5090OFFLINE00—0.0%
RECENT ROUNDS · NEWEST FIRST
ROUNDSTARTEDNODESCOMPUTESYNCSENTDDP WOULD SENDSAVEDLOSS
010-17 00:20 UTCRUNNING130m 00s0m 00s————
HOW DECENTRALIZED TRAINING WORKS · FOUR STATIONS

Train a lot. Talk rarely.

Decentralized training keeps the network quiet: miners only talk once per outer round, not after every step.
01 · INNER STEPSTrain locallyEach miner runs many optimizer steps on its own GPUs, with no network traffic in between.
02 · PSEUDO-GRADIENTMeasure the driftWhen the round ends, a miner's change is how far its weights moved from the shared model.
03 · OUTER SYNCSync once a roundMiners exchange pseudo-gradients once per outer round. That is a fraction of the bandwidth DDP needs.
04 · OUTER STEPNesterov updateAn outer Nesterov optimizer applies the combined update to the one shared model. Then the next round starts.
MOTIONcyan coin hops station to station · the sync tile blinks twice while the miners exchange pseudo-gradients · the outer-step tile flashes cream when the shared model updates
INSERT COIN

Hypertrain on Subnet 100.

The runs and rounds above are served by the Subnet 100 challenge. No reward has matured yet, so nothing has been paid. Bounty is open too.

OPEN /BOUNTY