IX2-0478
① SA Source
- Source: 開啟完整 SA 文章
- Section:
Optimizing Inference with Wide EP + Disaggregated Serving - Line hint:
690
Context Before
Source: SemiAnalysis
Evidence
As we move to slightly lower interactivities, batch sizes remain small enough that expert weights are still sharded via TP rather than EP
Context After
Source: SemiAnalysis InferenceX ↗
② Atomic Claim
當 interactivity 稍微下降時，batch size 仍偏小，因此 expert weights 仍較適合用 TP 分片，而非 EP。
- Epistemic Mode:
ASSERTED - Mapping Status:
COMPLETE
③ Semantic Frame
{
"additional_nodes": [],
"frame_type": "RELATION",
"object": {
"id": "04_knowledge_base/Tensor Parallelism",
"label": "TP"
},
"predicate": "PARTITIONS_INTO",
"qualifiers": {
"condition_text": null,
"numeric_mentions": [],
"temporal_mentions": []
},
"subject": {
"id": "04_knowledge_base/Batch size",
"label": "batch size"
}
}④ Canonical Entity Mapping
| Role | Surface Label | Canonical Target |
|---|---|---|
| subject | batch size | 04_knowledge_base/Batch size |
| object | TP | 04_knowledge_base/Tensor Parallelism |
⑤ Human Review
請在 Properties 逐項確認：
- 原文 → Atomic Claim 是否忠實
- Atomic Claim → Semantic Frame 是否忠實
- Canonical Entity mapping 是否正確
- Epistemic mode 是否保留原文語氣
- 最後選擇
review_action
Review state
Markdown 內文不是正式 approval。只有 Apply bridge 寫入的 Decision Ledger event 才是正式決策。