prime-grokking.git/tests, branch main
Prime Grokking — can minimal architectures grok next-prime? Pre-registered negative-class experiment (weight-tied RNN vs transformer).
http://git.jayrup.me/c/prime-grokking.git/atom?h=main
2026-08-29T22:19:29Z
implement Addendum 8: 6-arm token-space recurrence suite (arms A, B, C, D1, D2, D3), tests, and jobs/e8.csv
2026-08-29T22:19:29Z
CaptainJack2491
jayrupnakawala@gmail.com
2026-08-29T22:19:29Z
urn:sha1:7b0fddb02b089b82b8d12b2dcac17bf9527817da
fix(eval): correct sieve-rank indexing + test expectations
2026-08-20T16:02:15Z
Void Agent
void@jayrup.hermes
2026-08-20T16:02:15Z
urn:sha1:cf83937689bb30e2b5fa6e3efaa2f115016a030b
- _compute_sieve_rank: find LARGEST k (not smallest) whose signature
contains all error preds — fixes P5(4) vs P5(1) bug
- _sieve_rank_signature: iterate all small_primes (not just p²≤n) to
catch composites like 209=11*19 where 11²>209
- Test expectations updated for [101,200] range (209 outside it)
- is_prime task: P1→P3 (no sieve-rank for classification tasks)
48/48 green.
fix(eval): P5/P6 sieve-rank ladder replaces hardcoded FLAGGED_SIEVE_PREDS
2026-08-20T15:55:49Z
Void Agent
void@jayrup.hermes
2026-08-20T15:55:49Z
urn:sha1:f8995751de925571d4eec669ab7ab8b1c3d76cb9
- probe_report: dynamic lo/hi (no more hardcoded [101,200])
- main(): probe range derived from range_end ([range_end+1, range_end+1000])
- P6 = exact (no probe misses), P5(k) = errors match rank-k sieve signature
- _sieve_rank_signature: computes composites with all factors > p_k
- _compute_sieve_rank: identifies rank from model's error predictions
- build_report.py: probe_fig uses sieve rank for coloring
- Tests updated for P5/P6 codes (48/48 green)
perf(cuda): add AMP fp16 autocast with GradScaler and torch.compile reduce-overhead mode
2026-08-17T15:35:28Z
CaptainJack2491
jayrupnakawala@gmail.com
2026-08-17T15:35:28Z
urn:sha1:7166dc47c9bd764e2cd1d7c3d9e046cdc439b58c
feat(gpu): add CUDA support, chunked eval, bisect prime search, and scaling to 100k (48 tests green)
2026-08-17T15:15:28Z
CaptainJack2491
jayrupnakawala@gmail.com
2026-08-17T15:15:28Z
urn:sha1:66b8b77feb5fafd3696044ee33a80f26cb693a13
fix D2: integers-mode probe is N/A by construction (OOV probe inputs), functional regression test; D2 scored (int tokens collapse task: O2, 0-13% vs digits 17-87%); NOTES + report Phase 4 updated; 42 tests
2026-08-17T14:26:16Z
Void Agent
void@jayrup.hermes
2026-08-17T14:26:16Z
urn:sha1:535bb1dcf727e2273b956553c0491ec80ad8ec98
phase3: task_mode (is_prime), train_frac, lr_decay flags + tests (41 green); Addendum 5 locked; 11-job batch
2026-08-16T13:41:54Z
Void Agent
void@jayrup.hermes
2026-08-16T13:41:54Z
urn:sha1:751cbe0d93af1fcc69147d27ca8babd63105e4dc
speedup: batched eval (46x, sieve-stub verified), run_sweep orchestrator (2-way parallel, idempotent, summaries), E1 jobs; 36 tests
2026-08-14T23:00:41Z
Void Agent
void@jayrup.hermes
2026-08-14T23:00:41Z
urn:sha1:38d6553048808f6b53488894fdb4c83211590ad4
eval: P1 diagnostic prediction-based (flagged set +209), classifier regression suite (10 tests, stub ground truth); 34 tests green
2026-08-14T12:27:54Z
Void Agent
void@jayrup.hermes
2026-08-14T12:27:54Z
urn:sha1:39b8218a5fe660200a833c03b36afbde6c119467
fix codex BLOCKERs: global fixed layout invariance, min_steps off-by-one, integers EOS alias, rerun guard, eval prereg-literal codes + dual-checkpoint honesty; +7 regression tests
2026-08-14T12:21:03Z
Void Agent
void@jayrup.hermes
2026-08-14T12:21:03Z
urn:sha1:9fecbe58cc0750e9e39b261a672b0c6e85e6ed4e