btx-e003 R2 ledger final
143e6f5 verified - campaign_a frame-A campaign: A1 front-fusion PASSED (0.87-0.91 both variants/seeds, bit-exact off), A2 consumption REFUTED (0.009/0.010 vs 0.15 line, controls 0.000/0.0003), A3 gated; all arms incl. controls shipped
- campaign_b Campaign B v2 (InfoNCE mix) both seeds PASS; retro-erank corrected after overwrite catch (HF copies were the true v1): spread v1 180 > v2 139 (baseline 239) — target granularity governs spread, not loss family
- campaign_c btx-e003 R2 ledger final
- 1.05 MB frame prototype v0.1: byte-state arm beats index twin (digits 0.275 vs 0.045, twin has only 200 keys for 999 states); RSA says the codebook kept C ancestry; entry channel real vs failing random control
- 511 Bytes frame prototype v0.1: byte-state arm beats index twin (digits 0.275 vs 0.045, twin has only 200 keys for 999 states); RSA says the codebook kept C ancestry; entry channel real vs failing random control
- 4.39 MB frame prototype v0.1: byte-state arm beats index twin (digits 0.275 vs 0.045, twin has only 200 keys for 999 states); RSA says the codebook kept C ancestry; entry channel real vs failing random control
- 2.26 kB frame prototype v0.1: byte-state arm beats index twin (digits 0.275 vs 0.045, twin has only 200 keys for 999 states); RSA says the codebook kept C ancestry; entry channel real vs failing random control
- 174 Bytes frame prototype v0.1: byte-state arm beats index twin (digits 0.275 vs 0.045, twin has only 200 keys for 999 states); RSA says the codebook kept C ancestry; entry channel real vs failing random control
- 9.47 kB frame prototype v0.1: byte-state arm beats index twin (digits 0.275 vs 0.045, twin has only 200 keys for 999 states); RSA says the codebook kept C ancestry; entry channel real vs failing random control
- 8.53 kB frame prototype v0.1: byte-state arm beats index twin (digits 0.275 vs 0.045, twin has only 200 keys for 999 states); RSA says the codebook kept C ancestry; entry channel real vs failing random control
- 13.2 kB frame prototype v0.1: byte-state arm beats index twin (digits 0.275 vs 0.045, twin has only 200 keys for 999 states); RSA says the codebook kept C ancestry; entry channel real vs failing random control
- 7.56 kB v0.2: sweep retracts the fusion-gap reading (storage-position asymmetry: bert front-loads, t5 back-loads; student concat 0.954 > teacher); sequence entry: all 999 states, control still fails, embedding table byte-arbitrary
- 3.22 kB v0.2: sweep retracts the fusion-gap reading (storage-position asymmetry: bert front-loads, t5 back-loads; student concat 0.954 > teacher); sequence entry: all 999 states, control still fails, embedding table byte-arbitrary
- 4.01 kB v0.2: sweep retracts the fusion-gap reading (storage-position asymmetry: bert front-loads, t5 back-loads; student concat 0.954 > teacher); sequence entry: all 999 states, control still fails, embedding table byte-arbitrary
- 108 kB frame prototype v0.1: byte-state arm beats index twin (digits 0.275 vs 0.045, twin has only 200 keys for 999 states); RSA says the codebook kept C ancestry; entry channel real vs failing random control
- 545 Bytes v0.2: sweep retracts the fusion-gap reading (storage-position asymmetry: bert front-loads, t5 back-loads; student concat 0.954 > teacher); sequence entry: all 999 states, control still fails, embedding table byte-arbitrary
- 2.92 kB v0.2: sweep retracts the fusion-gap reading (storage-position asymmetry: bert front-loads, t5 back-loads; student concat 0.954 > teacher); sequence entry: all 999 states, control still fails, embedding table byte-arbitrary
