V6.7: upload scripts/test_bbpe_refit_serial.py (10.3KB) — FASE1+FASE2 training results

V6.7: upload scripts/fix_bbpe_refit_oom.py (19.6KB) — FASE1+FASE2 training results

V6.7: upload scripts/run_v67_fase2_only.py (8.2KB) — FASE1+FASE2 training results

V6.7: upload scripts/run_v67_train_quick.py (15.7KB) — FASE1+FASE2 training results

V6.7: scrub literal HF_TOKEN from upload script (security)

V6.7: deprecated → scripts/deprecated/upload_v6_5_7ds_attn_v3.py

V6.7: deprecated → scripts/deprecated/upload_v6_5_7ds_attn_v2.py

V6.7: deprecated → scripts/deprecated/upload_v6_5_7ds_attn.py

V6.7: deprecated → scripts/deprecated/upload_v6_5_7ds.py

V6.7: deprecated → scripts/deprecated/train_v6_5_7ds_attn_v3.py

V6.7: deprecated → scripts/deprecated/train_v6_5_7ds_attn_v2.py

V6.7: deprecated → scripts/deprecated/train_v6_5_7ds_attn.py

V6.7: deprecated → scripts/deprecated/train_v6_5_7ds.py

V6.7: deprecated → scripts/deprecated/train_v6_4.py

V6.7: upload scripts/som_auto_adjust_runner.py (som_auto_adjust_runner + train script V6.7)

V6.7: upload scripts/train_v6_5_v2.py (som_auto_adjust_runner + train script V6.7)

V6.5-V2-dynamic: upload batch (scripts + state + model) [80 files]

V7 FINAL: FASE1+FASE2 completed (10000 samples, 20 punishments)

V6.5-V4-canonical-256: buffer=256 + grid=(4,4,4,4)=256 + BBPE serial fix

V6.5-V3-hyp-synergy: SynergyHypothesisEnsemble (4 pilares: diversidade k-means++, cooperação, competição, ortogonalidade) + testes de validação (7 files)

V6.5-V2-dynamic: upload batch (scripts + state + model) [79 files]

V6.5-V2-dynamic: upload batch (scripts + state + model) [79 files]

V6.5-V2-parallel-reorganize: ParallelSOM4D + ReorganizingSOM4D + OOM-killer guards + unified state (75 files)

V6.5-V2-dynamic: upload batch (scripts + state + model) [79 files]

V6.5-V2-unified: cleanup misplaced root files + old scripts [18 files deleted] — removes duplicates from repo root (they live in src/bigru_t/...) and outdated scripts in scripts/ (now in deprecados/)

V6.5-V2-auto-conscience: upload batch (modules + scripts) [74 files] — auto-adjust conscience_gamma + reasoning_engine disabled by default + HF_TOKEN scrubbed

V6.5-V2-dynamic: upload batch (scripts + state + model) [79 files]

V6.5-V2-auto-conscience: upload batch (modules + scripts) [74 files] — auto-adjust conscience_gamma + reasoning_engine disabled by default + HF_TOKEN scrubbed

V6.5-V2-auto-conscience: upload batch (modules + scripts) [73 files] — auto-adjust conscience_gamma + reasoning_engine disabled by default + HF_TOKEN scrubbed

V6.5-V2-metrics-FIX-2: fix HF organization — remove 9 root-level duplicates, upload 13 files to correct src/bigru_t/ paths

Delete scripts/__pycache__/upload_v6_5_7ds_attn_v2.cpython-312.pyc with huggingface_hub

Delete scripts/__pycache__/train_v6_5_7ds_attn_v2.cpython-312.pyc with huggingface_hub

V6.5-attn-v2 cleanup: remove legacy scripts/upload_v6_5_7ds_attn_v2.py

V6.5-attn-v2 cleanup: remove legacy scripts/train_v6_5_7ds_attn_v2.py

V6.5-attn-v2 cleanup: remove legacy scripts/upload_v6_5_7ds_attn.py

V6.5-attn-v2 cleanup: remove legacy scripts/train_v6_5_7ds_attn.py

V6.5-attn-v2 cleanup: remove legacy scripts/upload_v6_5_7ds.py

V6.5-attn-v2 cleanup: remove legacy scripts/train_v6_5_7ds.py

V6.5-attn-v3: CORREÇÃO MATEMÁTICA do predict() — rótulos dinâmicos extraídos do label_registry (substitui 'gato'/'cachorro' fixos). NOVO: punish_during_inference() — mirror do protocolo de punição do treinamento (1ª → activate_hypothesis, 2ª → set_ewc_reference). META ≥6000 samples, attention ACTIVE_AND_FUNCTIONAL, SmoothQuant W8A8, 864 neurons, MTP K=6, 3 perguntas reais sem ajuda (Luva de Pedreiro Távila, Lula reserva valor, Amazonas força-tarefa vítimas) + punição na inferência via camada de hipótese. Removed legacy v2 reports/scripts (superseded by v3).

V6.5-attn-v2 cleanup: remove legacy scripts/upload_v6_5_7ds_attn.py

V6.5-attn-v2 cleanup: remove legacy scripts/train_v6_5_7ds_attn.py

V6.5-attn-v2 cleanup: remove legacy scripts/upload_v6_5_7ds.py

V6.5-attn-v2 cleanup: remove legacy scripts/train_v6_5_7ds.py

V6.5-attn-v2: META ≥6000 samples (6020 REAL), salvar estados do modelo, 3 perguntas reais sem ajuda (Luva de Pedreiro Távila, Lula reserva valor, Amazonas força-tarefa vítimas), attention ACTIVE_AND_FUNCTIONAL (6161 calls/0 err), SmoothQuant W8A8 (final err=0.0056), 864 neurons, MTP K=6, tool_coordinator workers reactivated, 11/11 verification PASS, reasoning GOOD (100% answer/think/plan/decompose), removed legacy v3/v4/v5/v6_1/v6_2/v6_3/v6_progressive + old v6_5_*/v6_5_final_*/v6_5_7ds_* + v6_5_7ds_attn_* v1 reports/scripts (superseded by v2)

V6.5-attn cleanup: remove legacy scripts/upload_v6_5_7ds.py

V6.5-attn cleanup: remove legacy scripts/train_v6_5_7ds.py

V6.5-attn: attention integrated to KLS (MultiHeadAttention n_heads=8), 100 samples/batch streaming, 5040 samples REAL (meta ≥5000), 864 neurons, SmoothQuant W8A8 (err=0.020), MTP K=6, tool_coordinator workers reactivated, 11/11 verification PASS, attention ACTIVE_AND_FUNCTIONAL, reasoning GOOD (100% answer/think), removed legacy v3/v4/v5/v6_1/v6_2/v6_3/v6_progressive + old v6_5_* + v6_5_7ds_* reports/scripts

V6.5-7ds cleanup: remove legacy scripts/xeon_runtime.py

V6.5-7ds cleanup: remove legacy scripts/upload_v6_5_resilient.py

V6.5-7ds cleanup: remove legacy scripts/upload_v6_4_resilient.py