kali20gakki/msAgent / skills/accuracy/rl-prefill-consistency-dump
rl-prefill-consistency-dump
Verified训推一致性(prefill)采集侧:为训练/推理两侧采集可比的 prefill dump(输入对齐 PROMPTS_ONLY → 参数对齐校验(dump 前前置:校验两侧 config 超参/结构/权重,不一致输出参数不一致点并对齐)→ msprobe 接入采集 → 采集验证 → 恢复代码)。按用户需求判断用法:① 仅采集 dump——用户只要 dump 数据,走采集主流程(0→7→9),直接产出两侧可比 dump;② 配合分析——用户要求找所有独立差异点/差异点隔离验证/配合 rl-consistency-dump-analysis 分析时,才按分析侧打桩指令用 msprobe `load` 覆盖边界二次采集(阶段 8)再恢复代码。差异分析、首差异定位、打桩指令、整体结论由 rl-consistency-dump-analysis 承担。触发词:训推一致、训推一致性、prefill 比对、msprobe dump/采集、精度比对数据采集、打桩二次采集、参数对齐校验、参数不一致点、mHC 配置不一致、权重不同步、清卡验环境后采数据;已采 dump 的分析转 rl-consistency-dump-analysis。
| File | Kind | Size |
|---|---|---|
| SKILL.md | skill md | 25.7 KB |
| references/msprobe-config.md | markdown | 5.4 KB |
| references/new-framework.md | markdown | 5.2 KB |
| references/param-alignment.md | markdown | 14.4 KB |
| references/pitfalls.md | markdown | 22.1 KB |
| references/prefill-logprob-diff.md | markdown | 5.6 KB |
| references/probs-diff-monitoring.md | markdown | 6.6 KB |
| references/slime.md | markdown | 5.4 KB |
| references/stubbing.md | markdown | 7.7 KB |
| references/task_card.json | config | 3.2 KB |
| references/templates/param_alignment_instruction.example.json | config | 1.4 KB |
| references/templates/stub_instruction.example.json | config | 1.2 KB |
| references/verl.md | markdown | 5.0 KB |
| references/vime.md | markdown | 9.5 KB |
| scripts/param_align_check.py | script | 8.7 KB |
| scripts/param_align_provider_dump.py | script | 5.7 KB |
| scripts/param_alignment_mapping.example.json | config | 2.3 KB |
references/task_card.json
View on GitHub{
"model": "<模型名>(<权重路径,如 xxx 诊断权重/完整权重>)",
"collection_scope": "确定性单样本 prefill (固定样本, seed 固定, TP=1, 关 shuffle/eval)",
"evidence_tool": "msprobe PrecisionDebugger (level=mix)",
"train_framework": "<训练框架,如 verl/slime/vime>",
"train_framework_version": "<版本>",
"train_backend": "<FSDP / Megatron / DeepSpeed / 自研>",
"architecture_mode": "<显式 micro_batch 循环 (FSDP 型, 包循环体) / 连续前向 (Megatron 型, 插 forward 内部)>",
"inference_engine": "<vLLM / SGLang / Transformers / 自研>",
"inference_engine_version": "<版本>",
"hardware": "<硬件后端,如 Ascend NPU> 卡 <M,N> (actor 1 + rollout 1)",
"stop_condition": "采到两侧可比的 prefill dump (level=mix)",
"stage3_onsite": {
"crop_location": "<现场 grep 定位的 PROMPTS_ONLY 裁剪位置,以现场符号为准>",
"instrument_location": "<现场定位的 log_prob 主前向接入位置;判断包循环体还是插 forward 内部>",
"main_forward_gate": "<主前向是否被缓存跳过(类似 can_reuse_log_probs_in_loss),采不到数据先怀疑它>",
"request_id": "<推理侧调度日志与训练 rollout 数据的对应字段;无现成日志就在接入处自己打>",
"padding": "<去训练端 padding 的开关/做法,两侧输入 shape 对齐真实 prompt 长度 (P18)>",
"infer_msprobe": "<引擎是否原生集成 msprobe(vLLM-ascend/SGLang 原生, 传 config 即启用);无原生集成按 new-framework.md §2 手写接入>"
},
"authorization": {
"AUTH_STOP_RUNNING_SERVICES": false,
"AUTH_EDIT_TARGET_CODE": true,
"AUTH_RUN_DUMP_JOB": true,
"AUTH_CLEAN_APPROVED_DIRS": true,
"AUTH_RESTORE_CODE": true,
"notes": "<按 SKILL 阶段 2 一次性问清并记录;停服务范围 / 允许改的代码 / 允许删的目录 / 采集后恢复,避免流程中反复打断>"
},
"environment": {
"ssh_target": null,
"container": "<容器标识,以现场为准(不预设容器形态,见框架线索卡对应说明)>",
"conda_env": "<环境激活方式>",
"workdir": "<工作目录,绝对路径>",
"log_path": "<日志位置>",
"artifact_root": "<会话/产物根目录>",
"credentials_note": "凭据不写入本卡或报告"
},
"assets": {
"model_path": "<权重路径>",
"baseline_command": "<基线命令,绝对路径>",
"train_msprobe_config": "<session>/train_dump_config.json (task=statistics, level=mix)",
"infer_msprobe_config": "<session>/infer_dump_config.json (task=statistics, level=mix)",
"train_dump_path": "<session>/dumps/train",
"infer_dump_path": "<session>/dumps/infer"
},
"constraints": {
"padding": "训练端 padding 必须去掉, 两侧输入 shape 对齐真实 prompt 长度 (P18)",
"determinism": "seed 固定, 关 shuffle, 单样本小 batch (num-rollout 1 / rollout-batch-size 1 / n-samples-per-prompt 1 / rollout-max-response-len 4 / rollout-temperature 0 / global-batch-size 1), enforce_eager",
"e2e_target": "response 段 E2E 指标(PROMPTS_ONLY 下为 0 值是'没算', 需补跑 PROMPTS_ONLY=0 标准 step 才有实测)",
"prefill_logprob_diff": "训练循环内直接算并打印训练日志 (主指标, mean≈0 是训推一致)"
}
}
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis. | Stars: 471k | Updated | Claude CodeCodex | Verified | Install |
| diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness. | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is available | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |