SA260324R-0132_r1

SA260324R-0132

① SA Source

• Source: 開啟完整 SA 文章
• Section: Speculative Decoding
• Line hint: 167

Context Before

For LPUs, deploying a draft model or MTP layers is quite different from applying AFD. FFNs are stateless, while draft models and MTP layers require dynamic KV cache loading. Each FFN is around hundreds of megabytes, whereas

Evidence

draft models and MTP layers take up tens of gigabytes.

Context After

To support this memory usage, LPUs can access up to 256 GB of DDR5 per Fabric Expansion Logic FPGAs on the LPX compute tray.

② Atomic Claim

draft model 與 MTP layer 會占用數十 GB 的容量。

• Epistemic Mode: ASSERTED
• Mapping Status: PARTIAL

③ Semantic Frame

{
  "frame_type": "CLAIM_ONLY",
  "qualifiers": {
    "condition_text": null,
    "numeric_mentions": [
      "tens of gigabytes"
    ],
    "temporal_mentions": []
  },
  "reason": "MIXED_MAPPED_AND_UNMAPPED_COMPONENTS_DRAFT_MODEL_HAS_NO_CANONICAL_ENTITY",
  "statement": "draft model 與 MTP layer 會占用數十 GB 的容量。"
}

④ Canonical Entity Mapping

Role	Surface Label	Canonical Target
—	—	—

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。