feat(sft): Phase 13 正式 7B SFT/LoRA baseline(launcher + evaluator + merge tool) - #11
Merged
Conversation
added 3 commits
August 19, 2026 21:38
…e launcher, verl LoRA->HF merge tool (phase 13)
…oxy baseline evaluator)
…ate, bounded checkpoint defaults, grad-accum divisibility guard, RL-init wording
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Phase 13 — 7B SFT baseline(reproducible baseline + candidate downstream-init artifact)
Commits
b8b5216feat(sft): 7B LoRA baseline launcher + choice-metrics evaluator + verl LoRA→HF merge tool85c1264feat(sft): baseline eval JSON 持久化每条 completion41c24e3refactor(sft): Phase-13 的end_to_end指标改名proxy_e2e(factorized/proxy 语义)e05f7edfix(sft): merge 前 review 修复(见下)Changed files
script/verl/sft/run_baseline.sh— Qwen2.5-7B-Instruct SFT/LoRA 训练 launcher(env 化,输出 hyperparams.json)script/verl/sft/evaluate_baseline.py— factorized/proxy 评估器(choice protocol,复用agent.evaluation)script/verl/sft/merge_lora_checkpoint.py— verl LoRA FSDP checkpoint → 合并 HF 目录Merge 前 review 修复(e05f7ed)
script.verl.sft.validate非 0 → 打印错误并exit 2,不再 warning 后继续训练(token-budget gate 逻辑不变)。SAVE_FREQ=50(低频)、MAX_CKPT_KEEP=4(上限,避免 7B FSDP ckpt ~15GB/个塞满盘);默认不绑定任何数据集步数。TRAIN_BATCH_SIZE不能被MICRO_BATCH_SIZE_PER_GPU × GPUS整除时直接报错退出,不再整数除法静默截断。验证
tests/task/ evaluation/ rl/→ 184 passed, 1 skippedbash -n run_baseline.sh/py_compile通过;grad-accum guard 功能测试(batch=7 → exit 2)通过proxy_e2e为 factorized 指标(Stage2 用预构造 gold bundle),true-E2E 见 PR feat(eval): Phase 13.5 classification framework freeze + true end-to-end evaluator #12说明
docs/reports/stage-13-sft-baseline-report.md(按仓库约定不入库)。