Training runs, benchmarks, deployments, dataset updates, data-provider failures — and, once the token exists, rewards and compute purchases. Everything here really happened.
It had stalled (a failed run (retrying)); the watchdog restarted that step.
14 Oct 16:50
Training started: Model Zero v0.68 (attempt 2)
Solana Launch Dataset v44 · boosted decision trees · Model Zero training server (CPU)
14 Oct 16:50
Retrying Model Zero v0.68
Attempt 2. The failed attempt stays in the history.
14 Oct 16:41
Model Zero v0.68 training FAILED
Reason: training process terminated before finishing (no final status written). Cost: $0.00. Awaiting a controlled retry.
14 Oct 15:53
Training started: Model Zero v0.68
Solana Launch Dataset v44 · boosted decision trees · Model Zero training server (CPU)
14 Oct 15:52
Model Zero v0.68 queued
Automatic search from v0.66: min samples leaf 40 → 10; min samples leaf 10 → 5, with automatic signal discovery.
14 Oct 15:50
Training completed: Model Zero v0.67
14:41:59 · trained on 163880 examples · $0 compute cost
14 Oct 01:08
Training started: Model Zero v0.67
Solana Launch Dataset v43 · boosted decision trees · Model Zero training server (CPU)
14 Oct 01:07
Model Zero v0.67 queued
Automatic search from v0.66: adds launch context (time, name style, launch tool track record), StonkFun and the market meta, with automatic signal discovery.
14 Oct 01:06
Training completed: Model Zero v0.66
11:41:25 · trained on 150276 examples · $0 compute cost
13 Oct 13:24
Training started: Model Zero v0.66
Solana Launch Dataset v42 · boosted decision trees · Model Zero training server (CPU)
13 Oct 13:23
Model Zero v0.66 queued
03-40-0.8--:discover — retrained on new data. Same configuration as v0.65 (best so far on the headline benchmark), retrained on the newest dataset.
13 Oct 13:22
Training completed: Model Zero v0.65
11:11:15 · trained on 150119 examples · $0 compute cost
13 Oct 02:10
Training started: Model Zero v0.65
Solana Launch Dataset v41 · boosted decision trees · Model Zero training server (CPU)
13 Oct 02:09
Model Zero v0.65 queued
Automatic search from v0.45: adds the market meta, with automatic signal discovery.
13 Oct 00:21
Model Zero v0.64 not trained
dataset too small: 93 snapshots on f10 against 168,694 for the live v0.45; that feature set covers too few launches yet.
13 Oct 00:08
Model Zero v0.64 queued
Automatic search from v0.45: adds who funded the early buyers, the creator's wallet and real on-chain holders, plus every newer signal; min samples leaf 40 → 5, with automatic signal discovery.
12 Oct 22:07
Model Zero v0.63 not trained
dataset too small: 93 snapshots on f10 against 168,694 for the live v0.45; that feature set covers too few launches yet.
12 Oct 22:07
Model Zero v0.63 queued
Automatic search from v0.45: adds who funded the early buyers, the creator's wallet and real on-chain holders, plus every newer signal; min samples leaf 40 → 5, with automatic signal discovery.
12 Oct 20:07
Model Zero v0.62 not trained
dataset too small: 93 snapshots on f10 against 168,694 for the live v0.45; that feature set covers too few launches yet.
12 Oct 20:07
Model Zero v0.62 queued
Automatic search from v0.45: adds who funded the early buyers, the creator's wallet and real on-chain holders, plus every newer signal; min samples leaf 40 → 5, with automatic signal discovery.
12 Oct 18:06
Model Zero v0.61 not trained
dataset too small: 93 snapshots on f10 against 168,694 for the live v0.45; that feature set covers too few launches yet.
12 Oct 18:06
Model Zero v0.61 queued
Automatic search from v0.45: adds who funded the early buyers, the creator's wallet and real on-chain holders, plus every newer signal; min samples leaf 40 → 5, with automatic signal discovery.
12 Oct 16:06
Model Zero v0.60 not trained
dataset too small: 93 snapshots on f10 against 168,694 for the live v0.45; that feature set covers too few launches yet.
12 Oct 16:06
Model Zero v0.60 queued
Automatic search from v0.45: adds who funded the early buyers, the creator's wallet and real on-chain holders, plus every newer signal; min samples leaf 40 → 5, with automatic signal discovery.
12 Oct 14:05
Model Zero v0.59 not trained
dataset too small: 66 snapshots on f10 against 168,694 for the live v0.45; that feature set covers too few launches yet.
12 Oct 14:05
Model Zero v0.59 queued
Automatic search from v0.45: adds who funded the early buyers, the creator's wallet and real on-chain holders, plus every newer signal; min samples leaf 40 → 5, with automatic signal discovery.
12 Oct 12:04
Model Zero v0.58 not trained
dataset too small: 66 snapshots on f10 against 168,694 for the live v0.45; that feature set covers too few launches yet.
12 Oct 12:04
Model Zero v0.58 queued
Automatic search from v0.45: adds who funded the early buyers, the creator's wallet and real on-chain holders, plus every newer signal; min samples leaf 40 → 5, with automatic signal discovery.
12 Oct 10:03
Model Zero v0.57 not trained
dataset too small: 66 snapshots on f10 against 168,694 for the live v0.45; that feature set covers too few launches yet.
12 Oct 10:03
Model Zero v0.57 queued
Automatic search from v0.45: adds who funded the early buyers, the creator's wallet and real on-chain holders, plus every newer signal; min samples leaf 40 → 5, with automatic signal discovery.
12 Oct 08:02
Model Zero v0.56 not trained
dataset too small: 66 snapshots on f10 against 168,694 for the live v0.45; that feature set covers too few launches yet.
12 Oct 08:02
Model Zero v0.56 queued
Automatic search from v0.45: adds who funded the early buyers, the creator's wallet and real on-chain holders, plus every newer signal; min samples leaf 40 → 5, with automatic signal discovery.
12 Oct 06:01
Model Zero v0.55 not trained
dataset too small: 66 snapshots on f10 against 168,694 for the live v0.45; that feature set covers too few launches yet.
12 Oct 06:01
Model Zero v0.55 queued
Automatic search from v0.45: adds who funded the early buyers, the creator's wallet and real on-chain holders, plus every newer signal; min samples leaf 40 → 5, with automatic signal discovery.
12 Oct 04:00
Model Zero v0.54 not trained
dataset too small: 66 snapshots on f10 against 168,694 for the live v0.45; that feature set covers too few launches yet.
12 Oct 04:00
Model Zero v0.54 queued
Automatic search from v0.45: adds who funded the early buyers, the creator's wallet and real on-chain holders, plus every newer signal; min samples leaf 40 → 5, with automatic signal discovery.
12 Oct 01:59
Model Zero v0.53 not trained
dataset too small: 66 snapshots on f10 against 168,694 for the live v0.45; that feature set covers too few launches yet.
12 Oct 01:59
Model Zero v0.53 queued
Automatic search from v0.45: adds who funded the early buyers, the creator's wallet and real on-chain holders, plus every newer signal; min samples leaf 40 → 5, with automatic signal discovery.
11 Oct 23:58
Model Zero v0.52 not trained
dataset too small: 66 snapshots on f10 against 168,694 for the live v0.45; that feature set covers too few launches yet.
11 Oct 23:58
Model Zero v0.52 queued
Automatic search from v0.45: adds who funded the early buyers, the creator's wallet and real on-chain holders, plus every newer signal; min samples leaf 40 → 5, with automatic signal discovery.
11 Oct 21:57
Model Zero v0.51 not trained
dataset too small: 66 snapshots on f10 against 168,694 for the live v0.45; that feature set covers too few launches yet.
11 Oct 21:57
Model Zero v0.51 queued
Automatic search from v0.45: adds who funded the early buyers, the creator's wallet and real on-chain holders, plus every newer signal; min samples leaf 40 → 5, with automatic signal discovery.
11 Oct 19:57
Model Zero v0.50 not trained
dataset too small: 66 snapshots on f10 against 168,694 for the live v0.45; that feature set covers too few launches yet.
11 Oct 19:57
Model Zero v0.50 queued
Automatic search from v0.45: adds who funded the early buyers, the creator's wallet and real on-chain holders, plus every newer signal; min samples leaf 40 → 5, with automatic signal discovery.
11 Oct 17:56
Model Zero v0.49 not trained
dataset too small: 66 snapshots on f10 against 168,694 for the live v0.45; that feature set covers too few launches yet.
11 Oct 17:56
Model Zero v0.49 queued
Automatic search from v0.45: adds who funded the early buyers, the creator's wallet and real on-chain holders, plus every newer signal; min samples leaf 40 → 5, with automatic signal discovery.
11 Oct 15:56
Model Zero v0.48 not trained
dataset too small: 66 snapshots on f10 against 168,694 for the live v0.45; that feature set covers too few launches yet.
11 Oct 15:56
Model Zero v0.48 queued
Automatic search from v0.45: adds who funded the early buyers, the creator's wallet and real on-chain holders, plus every newer signal; min samples leaf 40 → 5, with automatic signal discovery.
11 Oct 13:55
Model Zero v0.47 not trained
dataset too small: 66 snapshots on f10 against 168,694 for the live v0.45; that feature set covers too few launches yet.
11 Oct 13:55
Model Zero v0.47 queued
Automatic search from v0.45: adds who funded the early buyers, the creator's wallet and real on-chain holders, plus every newer signal; min samples leaf 40 → 5, with automatic signal discovery.
11 Oct 11:55
Training completed: Model Zero v0.46
00:00:13 · trained on 99 examples · $0 compute cost
11 Oct 11:55
Training started: Model Zero v0.46
Solana Launch Dataset v37 (f10) · boosted decision trees · Model Zero training server (CPU)
11 Oct 11:55
Model Zero v0.46 queued
Automatic search from v0.40: adds who funded the early buyers, the creator's wallet and real on-chain holders, plus every newer signal, with automatic signal discovery.
11 Oct 11:54
Training completed: Model Zero v0.45
09:53:30 · trained on 147480 examples · $0 compute cost
11 Oct 02:01
Training started: Model Zero v0.45
Solana Launch Dataset v36 · boosted decision trees · Model Zero training server (CPU)
11 Oct 02:00
Model Zero v0.45 restarted
It had stalled (dataset preparation); the watchdog restarted that step.
11 Oct 01:04
Model Zero v0.45 restarted
It had stalled (dataset preparation); the watchdog restarted that step.
11 Oct 00:00
Model Zero v0.45 restarted
It had stalled (dataset preparation); the watchdog restarted that step.
10 Oct 23:00
Model Zero v0.45 restarted
It had stalled (dataset preparation); the watchdog restarted that step.
10 Oct 22:00
Model Zero v0.45 restarted
It had stalled (dataset preparation); the watchdog restarted that step.
10 Oct 21:00
Model Zero v0.45 restarted
It had stalled (dataset preparation); the watchdog restarted that step.
10 Oct 20:07
Model Zero v0.45 restarted
It had stalled (dataset preparation); the watchdog restarted that step.
10 Oct 19:06
Model Zero v0.45 restarted
It had stalled (dataset preparation); the watchdog restarted that step.
10 Oct 18:00
Model Zero v0.45 restarted
It had stalled (dataset preparation); the watchdog restarted that step.
10 Oct 17:04
Model Zero v0.45 restarted
It had stalled (dataset preparation); the watchdog restarted that step.
10 Oct 16:04
Model Zero v0.45 restarted
It had stalled (dataset preparation); the watchdog restarted that step.
10 Oct 15:01
Model Zero v0.45 restarted
It had stalled (dataset preparation); the watchdog restarted that step.
10 Oct 14:00
Model Zero v0.45 restarted
It had stalled (dataset preparation); the watchdog restarted that step.
10 Oct 13:07
Model Zero v0.45 restarted
It had stalled (dataset preparation); the watchdog restarted that step.
10 Oct 12:01
Model Zero v0.45 restarted
It had stalled (dataset preparation); the watchdog restarted that step.
10 Oct 11:00
Model Zero v0.45 restarted
It had stalled (dataset preparation); the watchdog restarted that step.
10 Oct 10:00
Model Zero v0.45 restarted
It had stalled (dataset preparation); the watchdog restarted that step.
10 Oct 09:20
Model Zero v0.45 restarted
It had stalled (dataset preparation); the watchdog restarted that step.
10 Oct 08:56
Model Zero v0.45 queued
03-40-0.8:discover — retrained on new data. Same configuration as v0.40 (best so far on the headline benchmark), retrained on the newest dataset.
10 Oct 08:55
Training completed: Model Zero v0.44
05:25:12 · trained on 73851 examples · $0 compute cost
10 Oct 03:30
Model Zero v0.44 restarted
It had stalled (a failed run (retrying)); the watchdog restarted that step.
10 Oct 03:30
Training started: Model Zero v0.44 (attempt 12)
Solana Launch Dataset v35 · boosted decision trees · Model Zero training server (CPU)
10 Oct 03:30
Retrying Model Zero v0.44
Attempt 11. The failed attempt stays in the history.
10 Oct 03:29
Model Zero v0.44 training FAILED
Reason: training process terminated before finishing (no final status written). Cost: $0.00. Awaiting a controlled retry.
10 Oct 03:29
Model Zero v0.44 training FAILED
Reason: training process terminated before finishing (no final status written). Cost: $0.00. Awaiting a controlled retry.
09 Oct 21:52
Model Zero v0.44 restarted
It had stalled (a failed run (retrying)); the watchdog restarted that step.
09 Oct 21:52
Training started: Model Zero v0.44 (attempt 10)
Solana Launch Dataset v35 · boosted decision trees · Model Zero training server (CPU)
09 Oct 21:52
Training started: Model Zero v0.44 (attempt 10)
Solana Launch Dataset v35 · boosted decision trees · Model Zero training server (CPU)
09 Oct 21:52
Retrying Model Zero v0.44
Attempt 10. The failed attempt stays in the history.
09 Oct 21:52
Retrying Model Zero v0.44
Attempt 10. The failed attempt stays in the history.
09 Oct 21:41
Model Zero v0.44 training FAILED
Reason: training process terminated before finishing (no final status written). Cost: $0.00. Awaiting a controlled retry.
09 Oct 04:12
Model Zero v0.44 restarted
It had stalled (a failed run (retrying)); the watchdog restarted that step.
09 Oct 04:12
Training started: Model Zero v0.44 (attempt 9)
Solana Launch Dataset v35 · boosted decision trees · Model Zero training server (CPU)
09 Oct 04:12
Retrying Model Zero v0.44
Attempt 9. The failed attempt stays in the history.
09 Oct 04:05
Model Zero v0.44 training FAILED
Reason: training process terminated before finishing (no final status written). Cost: $0.00. Awaiting a controlled retry.
09 Oct 03:31
Model Zero v0.44 restarted
It had stalled (a failed run (retrying)); the watchdog restarted that step.
09 Oct 03:31
Training started: Model Zero v0.44 (attempt 8)
Solana Launch Dataset v35 · boosted decision trees · Model Zero training server (CPU)
09 Oct 03:31
Retrying Model Zero v0.44
Attempt 8. The failed attempt stays in the history.
09 Oct 03:26
Model Zero v0.44 training FAILED
Reason: training process terminated before finishing (no final status written). Cost: $0.00. Awaiting a controlled retry.
09 Oct 00:55
Model Zero v0.44 restarted
It had stalled (a failed run (retrying)); the watchdog restarted that step.
09 Oct 00:55
Training started: Model Zero v0.44 (attempt 7)
Solana Launch Dataset v35 · boosted decision trees · Model Zero training server (CPU)
09 Oct 00:55
Retrying Model Zero v0.44
Attempt 7. The failed attempt stays in the history.
09 Oct 00:43
Model Zero v0.44 training FAILED
Reason: training process terminated before finishing (no final status written). Cost: $0.00. Awaiting a controlled retry.
08 Oct 20:30
Model Zero v0.44 restarted
It had stalled (a failed run (retrying)); the watchdog restarted that step.