publish_model: take the checkpoint source from --ckpt-repo instead of a hard-coded repo name

publish_model.py: ship tokenizer_config.json + special_tokens_map.json (Gate 5 could not load a bare tokenizer.json)

publish_model.py: drop the local 'import shutil' that shadowed the module name and broke Gate 5

publish_model.py: drop max_workers (not in the Kaggle image's Hub client)

publish_model: clean room before the import, cache_dir, two more required card fields, gated != empty (E-046)

publish_model: anonymous clean room, every-file read-back, list-only-on-404 resume, card fields read from run_summary (review E-045)

review round 3 (probe/model-card/eval): rank-visible eval failure + cross-rank peak memory + complete run_summary; card derived-and-asserted from config; harness install --no-deps and task ids confirmed; torchrun redirects 3 + tee 3

Phase 5 tooling: publish the final weights with a card whose every number is read back off the Hub (mix manifest, audit.json, run_summary); parameter count recomputed from the safetensors header rather than asserted; batched commits; anonymous sha read-back; --verify-only is the Gate 5 clean-room load+generate

review round: pointer written+read-back before prune, verify retried, latest rolled to final, stale-pointer scan on resume, ddp_timeout 4h, chunked batched commits, manifest self-consistency gate, full content verification after upload

publish_mix: resume listing must read lfs.oid as a dict (getattr on a dict returned None for every shard, so a 'resume' re-sent all 2.3 GB); hard-exit 11/12 when a batched commit leaves a path absent or a remote sha that disagrees with the local bytes

Upload build/publish_mix.py with huggingface_hub

Upload build/publish_mix.py with huggingface_hub

Upload build/build_mix.py with huggingface_hub

Upload build/publish_mix.py with huggingface_hub

Upload build/build_mix.py with huggingface_hub

Upload build/audit_contamination.py with huggingface_hub

Upload build/publish_mix.py with huggingface_hub

Upload build/build_mix.py with huggingface_hub

Upload build/audit_contamination.py with huggingface_hub

Upload build/audit_contamination.py with huggingface_hub

Upload build/publish_mix.py with huggingface_hub

Upload build/build_mix.py with huggingface_hub

Upload build/audit_contamination.py with huggingface_hub

audit: vectorised rolling-hash grams + sorted-array lookup; the Python-set version could not finish 1.2B tokens

publish: ship tokenizer.json with the mix, sha-checked against the manifest

audit: composite AUDIT_PASSED verdict; missing shards are unmeasured, not clean

rehearsal: make the script self-contained re sys.path/cwd (E-021)

rehearsal: grab_json returns a reason, manifest fallback, single-line VERDICT

build_mix: no stray print inside BUILD_JSON block (E-019); manifest gains n_shards

download_missing must fetch record.json too (.bin-only filter broke restore); rehearsal pre-cleans stale test repo

add publish_mix: LFS-first upload, generated card + attribution, Hub-side verification

stage listing raises instead of returning {}; empty-repo case handled explicitly; verify refuses manifest-less PASS; +val holdout in merge

hubsync.ensure_repo must install creds first (real cause of the 401); correct E-017 diagnosis

build_mix: hub-init failure is fatal (not silent), empty merge exits 4; add token-scope probe

hub sync for staged sources + cold-resume rehearsal script

smoke PASSED (4/4 sources per shard, 40/40 boundaries); add contamination audit; raw target 1.23B = 23% headroom

build_mix v2: stage->merge interleave + doc offset table; verify_mix asserts source-mix floor

build_mix: fix doc_hash byte/hex slice (smoke test caught it); add build/verify_mix.py

Phase 2: build_mix -- resumable, cross-source shards, seeded intra-shuffle, mandatory en filter