$8.585Observed · provisional
0.682Public leaderboard
185of 5,391 teams
0.765Same ranked snapshot
-0.083Balanced accuracy
18:00Daily · America/New_York
01 / Capital ledger
USD · FAILED ATTEMPTS INCLUDED ONCE| Campaign / item | Observed $ | As of | Billing state | Source |
|---|---|---|---|---|
| Qwen one-pass model | 4.856 | 14:05 EST | Completed; provisional invoice rows | Receipt ↗ |
| Qwen failed hosts | 0.899 | 14:05 EST | Two failures included, not added again | Receipt ↗ |
| Qwen three-pass run | 2.292 | 14:05 EST | Cached accumulating invoice; later charges unknown, not freshly observed | Receipt ↗ |
| GLM CUDA pilot | 0.538 | 14:25 EST | Four attempts; rentals destroyed; provisional | Receipt ↗ |
QWEN LIMIT$8.047 / $24
GLM LIMIT$0.538 / $10
Cached invoices: Qwen 14:05 EST; GLM 14:25 EST. Later Qwen charges and settlement unknown. Limits are separate; not available account balances.
02 / Scored result tape
PUBLIC LEADERBOARD / NOT VALIDATION| Entry | Score | Date | Source |
|---|---|---|---|
| Previous public ensemble | 0.651 | 27 Oct | Receipt ↗ |
| Public reproduction | 0.681 | 10 Nov | Receipt ↗ |
| Matched public-only control | 0.682 | 10 Nov | Receipt ↗ |
Matched additions / vs 0.682 public control
| Member | Score | Δ |
|---|---|---|
| +10% old Qwen | 0.678 | -0.004 |
| +10% tokenizer | 0.675 | -0.007 |
| + token and byte | 0.669 | -0.013 |
Scored progression / zero-to-one scale
+0.031
Token + byte scored 0.669 at 16:13 EST / 21:13 UTC. Best remains 0.682. Public 0.683 control, full-pool and mixed candidates have no hidden score.
03 / Experiment blotter
NAMED BASELINES · SOURCE RECEIPTS · MEASURED OUTCOMES| # | Experiment | Evidence type | Result | Baseline → score | Interpretation / limits | Source |
|---|---|---|---|---|---|---|
| 01 | Full-pool Qwen | validation | +3.37 pp | 0.587169 → 0.620882 | Standalone: five prompt folds, one seed. Trainer and exposure changed together. | Receipt ↗ |
| 02 | Mistral diversity lead | validation | +1.78 pp | 0.622322 → 0.640127 | Older full-stack reference, cross-fitted blend weight; author-supplied predictions, not fully nested. Own full training staged. | Receipt ↗ |
| 03 | OLMo diversity lead | validation | +0.89 pp | 0.622322 → 0.631228 | Same older reference; author-supplied predictions and cross-fitted weight. Interval crosses zero; own weights not reproduced. | Receipt ↗ |
| 04 | Liquid frozen encoder | validation | −0.74 pp | 0.543060 → 0.535650 | 5% blend on old public reference; exact checkpoint and all 2,699 rows. Full-backbone training is a separate experiment. | Receipt ↗ |
| 05 | Pruned GLM | engineering | CUDA works | — | 45 layers, 62 exact shard hashes. Three passages at 0.743–0.794 seconds; 36.80 GiB peak reserved. First eight experts, not calibrated. No classifier accuracy yet. | Receipt ↗ |
| 06 | Qwen three-pass | progress | 14,000 / 44,851 | — | Latest received recovery recorded in current main queue: step 14,000. Remote process not checked by reporting task; no completed validation score. | Receipt ↗ |
| 07 | Liquid supervised encoder | validation | +0.56 pp | 0.543060 → 0.548615 | Seed 29, five prompt folds, explored 5% blend on older public reference. Not a board score or three-seed result. Seed 43 first fold positive; local queue process verified alive at daily review, seed 43 fold 2 logged started. | Receipt ↗ |
No experiments match. Clear search or select ALL.
04 / Next bets
RESEARCH QUEUE / NOT NEW SPEND AUTHORIZATION- 01Score matched 0.683 control and full-pool Qwen when capacity permits.
- 02Verify full-inference Gemini / Mistral candidates; previews are not scores.
- 03Prove GLM on dual T4, then calibrate experts on competition prose.
- 04Finish Liquid seeds and verify creative specialist weights.
05 / Accounting & evidence notes
READ THE TIMESTAMPSSelected current rental campaigns and research results — not an all-time account reconciliation.
Separate historical API usage: $6.9033 + $0.1963 reserve. Receipt ↗
Basis, exclusions & comparison limits
Separate historical API project; not included in the rental subtotal. Earlier pilot, local electricity, subscriptions, free Kaggle quota and other older rentals are not reconciled here.
- Percentage-point validation gains use different named baselines. Never add them or compare them with leaderboard gains.
- Validation shares explored on shared prompts; public-model training exposure not independently established.
- No all-time spend total is claimed. Historical API usage is separate; other older work is not reconciled.
- Project budget caps are separate limits, not money spent or a fungible account balance.
- Daily review: token-plus-byte failed to improve: 0.669 versus 0.682 (receipt 16:13 EST); rank remains cached at 15:32 EST.
- Creative collection saved receipts as of 2026-11-10T23:00:18.157862+00:00: returned usage $6.051409, unknown-charge reserve $0.175014, separate $10 cap. Includes failed attempts; not in rental subtotal and not a settled invoice. See reports/scorecard/creative-usage.json.
- Creative GLM/Kimi lipogram: +0.000738 at 30% within-family blend on older reference, three-seed mean; bootstrap interval crosses zero. All-creative +0.000368; standalone declined. Not a leaderboard gain.
- Generated Gemini: +0.001481; GLM/Kimi: +0.000373 at 30% within-family on older reference. Gemini standalone fell 0.015. Validation only; full ensemble candidates remain unscored.
- Evidence source gap: remote Qwen/creative process state and fresh provider invoices were not queried. Saved receipts and current main reviewed; no Kaggle requests.