evidence_class string | execution dict | opd dict | optimizer dict | path_source string | precision string | prompts_path string | rollout dict | run_id string | seed int64 | source_commit string | student dict | teacher dict | timing_scope dict | tracking dict |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
implementation_validation | {
"gpu_count": 1,
"placement": "teacher-student-colocated-sequential",
"student_device_index": 0,
"teacher_device_index": 0
} | {
"clip_ratio_c": 10,
"clip_ratio_high": 0.28,
"clip_ratio_low": 0.2,
"large_chunk_threshold": 6,
"loss_agg_mode": "token-mean",
"loss_max_clamp": 10,
"objective": "semantic-prior-clipped-pg",
"variant": "upstream-code"
} | {
"gradient_clip": 1,
"learning_rate": 0.000001,
"name": "AdamW",
"steps": 10,
"warmup_steps": 10,
"weight_decay": 0.01
} | measured: canary-config.json validated on copd, confirmed by Block S15/S18 | bfloat16 | /opt/dpca/reproduction/prompts/dpca_10step_prompts.json | {
"max_new_tokens": 64,
"min_new_tokens": 16,
"temperature": 1,
"top_k": 0,
"top_p": 1
} | dpca-node-copd-0-0-1-1xb200-colocated-tb-10step-20261006-r2 | 44 | 927a8264f2e303b7f82c2d331a58fd4240c8805a | {
"container_path": "/workspace/storage-shared/nlp/tungks/dpca/cache/models/student_instruct",
"repo_id": "unsloth/Meta-Llama-3.1-8B-Instruct",
"revision": "a2856192dd7c25b842431f39c179a6c2c2f627d1"
} | {
"container_path": "/workspace/storage-shared/nlp/tungks/dpca/cache/models/teacher",
"repo_id": "Qwen/Qwen3-8B",
"revision": "b968826d9c46dd6066d109eabc6255188de91218"
} | {
"actual_global_batch_size": 1,
"actual_prompt_source": "ten-versioned-arithmetic-prompts",
"actual_response_cap": 64,
"comparison_run_id": "dpca-paper8b-timing-2xb200-10step-20260906-r1",
"microbenchmark": true,
"not_an_efficacy_run": true,
"paper_global_batch_size": 128,
"paper_response_cap": 16384,
... | {
"accelerator": "b200x1",
"backend": "tensorboard",
"method": "dpca",
"platform": "copd",
"regime": "on-policy",
"wandb": false
} |
YAML Metadata Warning:empty or missing yaml metadata in repo card
Check out the documentation for more information.
dpca-b200-source
Tooling that builds, qualifies, packages and publishes the portable B200 runtime for
the DPCA reproduction of On-Policy Distillation (ivanniu/On-Policy-Distill).
This repo is the harness, not the runtime. The runtime payload (venv, source tree, node helpers) and the model weights live in a separate dataset repo, because they are ~56 GB and the tooling is 61 KB.
What is here
| File | What it is |
|---|---|
dpca-b200-source-20261005.tar.gz |
The tooling tree: modal/, node/, tools/, tests/, evidence/, PROJECT.md |
dpca-b200-source-20261005.SHA256SUMS |
sha256 of every file inside the archive, as extracted |
Verify before using
sha256sum dpca-b200-source-20261005.tar.gz
# 99faca25dd4c5f4660411d3f3ac3278a1fd00561af61e28366b9caddb0cd5ea9
tar -xzf dpca-b200-source-20261005.tar.gz
sha256sum -c dpca-b200-source-20261005.SHA256SUMS
Provenance
| Upstream paper repo | https://github.com/ivanniu/On-Policy-Distill |
| Upstream commit | 927a8264f2e303b7f82c2d331a58fd4240c8805a |
| Overlay branch | repro/modal-dpca-smoke |
| Overlay commit | fc0a34b216e29f8b12e99794731e3ebaa747c6d8 |
| Overlay fork | https://github.com/sontungkieu/On-Policy-Distill |
The overlay is 4 commits on top of the upstream commit. That count is a build-time
assertion (OVERLAY_COMMITS_AHEAD = 4), not a claim in this README.
Scope, honestly stated
The tooling proves the runtime executes the paper's objective on the paper's Table 4 model pair, on B200, from a portable payload that another machine can verify byte-for-byte. It does not reproduce the paper's numbers: the paper uses 16 GPUs, 500 steps, response length 16384 and global batch 128. The student SFT checkpoint the paper starts from was never published, and the vLLM rollout path was never qualified.
PROJECT.md in the archive carries the full record, including the defects that only
appeared on a second machine.
License
Apache-2.0, matching the upstream repository.
- Downloads last month
- 77