Datasets:
Tasks:
Text Retrieval
Modalities:
Text
Formats:
json
Sub-tasks:
document-retrieval
Languages:
English
Size:
1K - 10K
License:
| {"answer": "96 hours; after the second failed retry; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-1caa28a4798830b5", "meaning-doc-befd2d72c3171109", "meaning-doc-351efbb1715e4502"], "hard_negative_texts": ["Cedar privacy requests memo 0: The payment hold is capped at 96 hours during Q1, for every request regardless of source. Archived exceptions are counted as active for audit sampling.", "Cedar privacy requests memo 0: The payment hold is capped at 94 hours during Q1, after the second failed retry. The note also says archived exceptions remain active.", "Cedar privacy requests memo 0: The payment hold target is at least 96 hours during Q1, after the second failed retry. Teams may exceed it when capacity is low."], "id": "meaning-q-2c42686e876a5257", "positive_doc_id": "meaning-doc-1cb6dd6fb4a2130d", "positive_text": "Cedar privacy requests memo 0: The payment hold is capped at 96 hours during Q1, after the second failed retry. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Cedar's privacy requests payment hold has a 96 hours cap in Q1 after the second failed retry, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "7 requests; after the second failed retry; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-a0a67bf1475549f4", "meaning-doc-421b2e5363128702", "meaning-doc-a029ba4fa2a18ac8"], "hard_negative_texts": ["Cedar data retention memo 1: The payment hold is capped at 7 requests during Q3, only when the owner is the platform team. Archived exceptions are counted as active for audit sampling.", "Cedar data retention memo 1: The payment hold is capped at 10 requests during Q3, after the second failed retry. The note also says archived exceptions remain active.", "Cedar data retention memo 1: The payment hold target is at least 7 requests during Q3, after the second failed retry. Teams may exceed it when capacity is low."], "id": "meaning-q-9f625920dfd4e5c0", "positive_doc_id": "meaning-doc-94fe659e68f032ad", "positive_text": "Cedar data retention memo 1: The payment hold is capped at 7 requests during Q3, after the second failed retry. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Cedar's data retention payment hold has a 7 requests cap in Q3 after the second failed retry, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "72 minutes; unless the owner is the platform team; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-4d2e1cb8b3f704c3", "meaning-doc-2e1dc1a72902229d", "meaning-doc-d9a92a54a7ddc9ef"], "hard_negative_texts": ["Cedar security review memo 2: The approval threshold is capped at 72 minutes during Q3, after the regional handoff is signed. Archived exceptions are counted as active for audit sampling.", "Cedar security review memo 2: The approval threshold is capped at 75 minutes during Q3, unless the owner is the platform team. The note also says archived exceptions remain active.", "Cedar security review memo 2: The approval threshold target is at least 72 minutes during Q3, unless the owner is the platform team. Teams may exceed it when capacity is low."], "id": "meaning-q-d0fc89caf7c2fb65", "positive_doc_id": "meaning-doc-ece9c42a86c77ac8", "positive_text": "Cedar security review memo 2: The approval threshold is capped at 72 minutes during Q3, unless the owner is the platform team. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Cedar's security review approval threshold has a 72 minutes cap in Q3 unless the owner is the platform team, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "96 days; only when the request is marked external; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-aab7fed0ea468874", "meaning-doc-035a81b559bd1b45", "meaning-doc-1ec42b7cdf16365d"], "hard_negative_texts": ["Basil security review memo 3: The review level is capped at 96 days during Q4, when the customer tier is starter. Archived exceptions are counted as active for audit sampling.", "Basil security review memo 3: The review level is capped at 98 days during Q4, only when the request is marked external. The note also says archived exceptions remain active.", "Basil security review memo 3: The review level target is at least 96 days during Q4, only when the request is marked external. Teams may exceed it when capacity is low."], "id": "meaning-q-669c18c6d9ceb242", "positive_doc_id": "meaning-doc-6240a98eb98afcbe", "positive_text": "Basil security review memo 3: The review level is capped at 96 days during Q4, only when the request is marked external. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Basil's security review review level has a 96 days cap in Q4 only when the request is marked external, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "18 days; after the second failed retry; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-29f6237649d4ca95", "meaning-doc-915a7f5e5589123a", "meaning-doc-ec2b413e5bab7bd3"], "hard_negative_texts": ["Basil clinical triage memo 4: The error budget is capped at 18 days during Q4, when the customer tier is starter. Archived exceptions are counted as active for audit sampling.", "Basil clinical triage memo 4: The error budget is capped at 20 days during Q4, after the second failed retry. The note also says archived exceptions remain active.", "Basil clinical triage memo 4: The error budget target is at least 18 days during Q4, after the second failed retry. Teams may exceed it when capacity is low."], "id": "meaning-q-4c9f04173106181f", "positive_doc_id": "meaning-doc-4cd77d54ba799cd3", "positive_text": "Basil clinical triage memo 4: The error budget is capped at 18 days during Q4, after the second failed retry. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Basil's clinical triage error budget has a 18 days cap in Q4 after the second failed retry, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "36 tickets; when the customer tier is enterprise; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-c2dd56cb7c7b402d", "meaning-doc-63f13065333692ea", "meaning-doc-efd05e62de82551e"], "hard_negative_texts": ["Kepler payments memo 5: The payment hold is capped at 36 tickets during Q1, only when the owner is the platform team. Archived exceptions are counted as active for audit sampling.", "Kepler payments memo 5: The payment hold is capped at 42 tickets during Q1, when the customer tier is enterprise. The note also says archived exceptions remain active.", "Kepler payments memo 5: The payment hold target is at least 36 tickets during Q1, when the customer tier is enterprise. Teams may exceed it when capacity is low."], "id": "meaning-q-f9917043e0bf1f61", "positive_doc_id": "meaning-doc-212e6f3acaa32c60", "positive_text": "Kepler payments memo 5: The payment hold is capped at 36 tickets during Q1, when the customer tier is enterprise. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Kepler's payments payment hold has a 36 tickets cap in Q1 when the customer tier is enterprise, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "18 percent; after the second failed retry; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-02cd5527ba82cc48", "meaning-doc-e6d463c34bc7ce22", "meaning-doc-80a07ea5b644559e"], "hard_negative_texts": ["Cedar release management memo 6: The retention period is capped at 18 percent during Q3, after the first successful retry. Archived exceptions are counted as active for audit sampling.", "Cedar release management memo 6: The retention period is capped at 15 percent during Q3, after the second failed retry. The note also says archived exceptions remain active.", "Cedar release management memo 6: The retention period target is at least 18 percent during Q3, after the second failed retry. Teams may exceed it when capacity is low."], "id": "meaning-q-51b0ec70c5a3374b", "positive_doc_id": "meaning-doc-319935123d25728c", "positive_text": "Cedar release management memo 6: The retention period is capped at 18 percent during Q3, after the second failed retry. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Cedar's release management retention period has a 18 percent cap in Q3 after the second failed retry, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "7 percent; after the second failed retry; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-96081d25ee094bb2", "meaning-doc-1609ead7dcfa0161", "meaning-doc-ff7dcd210f10bd5a"], "hard_negative_texts": ["Lumen data retention memo 7: The sampling rate is capped at 7 percent during Q2, for every request regardless of source. Archived exceptions are counted as active for audit sampling.", "Lumen data retention memo 7: The sampling rate is capped at 9 percent during Q2, after the second failed retry. The note also says archived exceptions remain active.", "Lumen data retention memo 7: The sampling rate target is at least 7 percent during Q2, after the second failed retry. Teams may exceed it when capacity is low."], "id": "meaning-q-0787509db645d64b", "positive_doc_id": "meaning-doc-d7719615ba1fdb91", "positive_text": "Lumen data retention memo 7: The sampling rate is capped at 7 percent during Q2, after the second failed retry. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Lumen's data retention sampling rate has a 7 percent cap in Q2 after the second failed retry, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "36 minutes; when the customer tier is enterprise; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-e51f3b944b143fa8", "meaning-doc-9fe0041ab027cbd9", "meaning-doc-54f66a8e2cffd28a"], "hard_negative_texts": ["Basil release management memo 8: The approval threshold is capped at 36 minutes during Q2, for every request regardless of source. Archived exceptions are counted as active for audit sampling.", "Basil release management memo 8: The approval threshold is capped at 38 minutes during Q2, when the customer tier is enterprise. The note also says archived exceptions remain active.", "Basil release management memo 8: The approval threshold target is at least 36 minutes during Q2, when the customer tier is enterprise. Teams may exceed it when capacity is low."], "id": "meaning-q-dd78a790212a5cf3", "positive_doc_id": "meaning-doc-ede68fe431bb5062", "positive_text": "Basil release management memo 8: The approval threshold is capped at 36 minutes during Q2, when the customer tier is enterprise. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Basil's release management approval threshold has a 36 minutes cap in Q2 when the customer tier is enterprise, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "12 tickets; after the second failed retry; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-f4d386a8d77a0d8b", "meaning-doc-33dee803b5979e95", "meaning-doc-768822606cece527"], "hard_negative_texts": ["Lumen privacy requests memo 9: The cache lifetime is capped at 12 tickets during Q3, when the customer tier is starter. Archived exceptions are counted as active for audit sampling.", "Lumen privacy requests memo 9: The cache lifetime is capped at 14 tickets during Q3, after the second failed retry. The note also says archived exceptions remain active.", "Lumen privacy requests memo 9: The cache lifetime target is at least 12 tickets during Q3, after the second failed retry. Teams may exceed it when capacity is low."], "id": "meaning-q-150965e13488196f", "positive_doc_id": "meaning-doc-8e7926e122b37dd9", "positive_text": "Lumen privacy requests memo 9: The cache lifetime is capped at 12 tickets during Q3, after the second failed retry. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Lumen's privacy requests cache lifetime has a 12 tickets cap in Q3 after the second failed retry, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "24 percent; unless the owner is the platform team; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-e9d0ba7a3a6d847b", "meaning-doc-9d295edc30ebedb3", "meaning-doc-43adfccb0165c66c"], "hard_negative_texts": ["Juniper edge caching memo 10: The payment hold is capped at 24 percent during Q4, when the customer tier is starter. Archived exceptions are counted as active for audit sampling.", "Juniper edge caching memo 10: The payment hold is capped at 26 percent during Q4, unless the owner is the platform team. The note also says archived exceptions remain active.", "Juniper edge caching memo 10: The payment hold target is at least 24 percent during Q4, unless the owner is the platform team. Teams may exceed it when capacity is low."], "id": "meaning-q-aced6c00a7deac88", "positive_doc_id": "meaning-doc-6a1e0c70f25de37c", "positive_text": "Juniper edge caching memo 10: The payment hold is capped at 24 percent during Q4, unless the owner is the platform team. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Juniper's edge caching payment hold has a 24 percent cap in Q4 unless the owner is the platform team, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "18 days; only when the request is marked external; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-c6b9947bcf1db167", "meaning-doc-db587ffbbfa496ae", "meaning-doc-9bee26a88a044bb4"], "hard_negative_texts": ["Lumen clinical triage memo 11: The sampling rate is capped at 18 days during Q3, after the first successful retry. Archived exceptions are counted as active for audit sampling.", "Lumen clinical triage memo 11: The sampling rate is capped at 15 days during Q3, only when the request is marked external. The note also says archived exceptions remain active.", "Lumen clinical triage memo 11: The sampling rate target is at least 18 days during Q3, only when the request is marked external. Teams may exceed it when capacity is low."], "id": "meaning-q-14bb49b73af1f903", "positive_doc_id": "meaning-doc-b6bdcb354f45c4c7", "positive_text": "Lumen clinical triage memo 11: The sampling rate is capped at 18 days during Q3, only when the request is marked external. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Lumen's clinical triage sampling rate has a 18 days cap in Q3 only when the request is marked external, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "72 hours; before the regional handoff is signed; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-5b91efd5e4a16eae", "meaning-doc-73ed63e9b590bb24", "meaning-doc-4a765acb8e574cb7"], "hard_negative_texts": ["Basil security review memo 12: The review level is capped at 72 hours during Q4, after the regional handoff is signed. Archived exceptions are counted as active for audit sampling.", "Basil security review memo 12: The review level is capped at 70 hours during Q4, before the regional handoff is signed. The note also says archived exceptions remain active.", "Basil security review memo 12: The review level target is at least 72 hours during Q4, before the regional handoff is signed. Teams may exceed it when capacity is low."], "id": "meaning-q-626cb063daec6b01", "positive_doc_id": "meaning-doc-7046641939de102c", "positive_text": "Basil security review memo 12: The review level is capped at 72 hours during Q4, before the regional handoff is signed. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Basil's security review review level has a 72 hours cap in Q4 before the regional handoff is signed, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "7 days; before the regional handoff is signed; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-cc0464b9bbcc8ed3", "meaning-doc-8ee57889372cd939", "meaning-doc-2a5fbd946671906e"], "hard_negative_texts": ["Kepler security review memo 13: The retention period is capped at 7 days during Q2, for every request regardless of source. Archived exceptions are counted as active for audit sampling.", "Kepler security review memo 13: The retention period is capped at 13 days during Q2, before the regional handoff is signed. The note also says archived exceptions remain active.", "Kepler security review memo 13: The retention period target is at least 7 days during Q2, before the regional handoff is signed. Teams may exceed it when capacity is low."], "id": "meaning-q-9c19bc5c201b5f57", "positive_doc_id": "meaning-doc-109427e1acf11297", "positive_text": "Kepler security review memo 13: The retention period is capped at 7 days during Q2, before the regional handoff is signed. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Kepler's security review retention period has a 7 days cap in Q2 before the regional handoff is signed, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "12 days; unless the owner is the platform team; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-3d5e3327cdfe3359", "meaning-doc-30c1a087bb5ec7eb", "meaning-doc-7f7e250e1835e0fa"], "hard_negative_texts": ["Fjord release management memo 14: The cache lifetime is capped at 12 days during Q2, when the customer tier is starter. Archived exceptions are counted as active for audit sampling.", "Fjord release management memo 14: The cache lifetime is capped at 18 days during Q2, unless the owner is the platform team. The note also says archived exceptions remain active.", "Fjord release management memo 14: The cache lifetime target is at least 12 days during Q2, unless the owner is the platform team. Teams may exceed it when capacity is low."], "id": "meaning-q-2a5125c6af083e28", "positive_doc_id": "meaning-doc-c818ccfeae0d3dec", "positive_text": "Fjord release management memo 14: The cache lifetime is capped at 12 days during Q2, unless the owner is the platform team. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Fjord's release management cache lifetime has a 12 days cap in Q2 unless the owner is the platform team, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "24 tickets; after the second failed retry; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-75b06098f08574e7", "meaning-doc-7c8c15c154105bf9", "meaning-doc-7ecbf48d8a3ba905"], "hard_negative_texts": ["Fjord edge caching memo 15: The payment hold is capped at 24 tickets during Q3, after the first successful retry. Archived exceptions are counted as active for audit sampling.", "Fjord edge caching memo 15: The payment hold is capped at 21 tickets during Q3, after the second failed retry. The note also says archived exceptions remain active.", "Fjord edge caching memo 15: The payment hold target is at least 24 tickets during Q3, after the second failed retry. Teams may exceed it when capacity is low."], "id": "meaning-q-29c92422bd528b09", "positive_doc_id": "meaning-doc-667ca081762adeaa", "positive_text": "Fjord edge caching memo 15: The payment hold is capped at 24 tickets during Q3, after the second failed retry. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Fjord's edge caching payment hold has a 24 tickets cap in Q3 after the second failed retry, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "72 minutes; after the second failed retry; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-2473562061544156", "meaning-doc-088d566785667c8f", "meaning-doc-0a186a77ac451aad"], "hard_negative_texts": ["Aurora data retention memo 16: The cache lifetime is capped at 72 minutes during Q3, for every request regardless of source. Archived exceptions are counted as active for audit sampling.", "Aurora data retention memo 16: The cache lifetime is capped at 70 minutes during Q3, after the second failed retry. The note also says archived exceptions remain active.", "Aurora data retention memo 16: The cache lifetime target is at least 72 minutes during Q3, after the second failed retry. Teams may exceed it when capacity is low."], "id": "meaning-q-96e4c0fb458dc1a1", "positive_doc_id": "meaning-doc-5f3d39846c5f2554", "positive_text": "Aurora data retention memo 16: The cache lifetime is capped at 72 minutes during Q3, after the second failed retry. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Aurora's data retention cache lifetime has a 72 minutes cap in Q3 after the second failed retry, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "24 hours; only when the request is marked external; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-52824418ea055dcb", "meaning-doc-5f01bcae5628b063", "meaning-doc-1238dfc607bfbcfb"], "hard_negative_texts": ["Cedar cloud billing memo 17: The error budget is capped at 24 hours during Q4, after the first successful retry. Archived exceptions are counted as active for audit sampling.", "Cedar cloud billing memo 17: The error budget is capped at 30 hours during Q4, only when the request is marked external. The note also says archived exceptions remain active.", "Cedar cloud billing memo 17: The error budget target is at least 24 hours during Q4, only when the request is marked external. Teams may exceed it when capacity is low."], "id": "meaning-q-6c1e5795031167ac", "positive_doc_id": "meaning-doc-18621fe31cfb50cb", "positive_text": "Cedar cloud billing memo 17: The error budget is capped at 24 hours during Q4, only when the request is marked external. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Cedar's cloud billing error budget has a 24 hours cap in Q4 only when the request is marked external, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "72 days; unless the owner is the platform team; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-6e721907f1226f14", "meaning-doc-a2b9682d711676a8", "meaning-doc-27056bbe75e4f61d"], "hard_negative_texts": ["Indigo payments memo 18: The approval threshold is capped at 72 days during Q1, when the customer tier is starter. Archived exceptions are counted as active for audit sampling.", "Indigo payments memo 18: The approval threshold is capped at 70 days during Q1, unless the owner is the platform team. The note also says archived exceptions remain active.", "Indigo payments memo 18: The approval threshold target is at least 72 days during Q1, unless the owner is the platform team. Teams may exceed it when capacity is low."], "id": "meaning-q-645ff74d36e6b142", "positive_doc_id": "meaning-doc-1715d58110f3632c", "positive_text": "Indigo payments memo 18: The approval threshold is capped at 72 days during Q1, unless the owner is the platform team. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Indigo's payments approval threshold has a 72 days cap in Q1 unless the owner is the platform team, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "18 hours; only when the request is marked external; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-dfb69770a193345e", "meaning-doc-a32eb92fb108dd77", "meaning-doc-e16f9f4d8eee6d34"], "hard_negative_texts": ["Lumen cloud billing memo 19: The sampling rate is capped at 18 hours during Q2, after the first successful retry. Archived exceptions are counted as active for audit sampling.", "Lumen cloud billing memo 19: The sampling rate is capped at 24 hours during Q2, only when the request is marked external. The note also says archived exceptions remain active.", "Lumen cloud billing memo 19: The sampling rate target is at least 18 hours during Q2, only when the request is marked external. Teams may exceed it when capacity is low."], "id": "meaning-q-a8c079362004ae8d", "positive_doc_id": "meaning-doc-3fbd4362b259c3e2", "positive_text": "Lumen cloud billing memo 19: The sampling rate is capped at 18 hours during Q2, only when the request is marked external. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Lumen's cloud billing sampling rate has a 18 hours cap in Q2 only when the request is marked external, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "12 requests; before the regional handoff is signed; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-542d8020f4b5e939", "meaning-doc-e6a6db894741b846", "meaning-doc-d22db5b2024fb90a"], "hard_negative_texts": ["Kepler payments memo 20: The sampling rate is capped at 12 requests during Q3, only when the owner is the platform team. Archived exceptions are counted as active for audit sampling.", "Kepler payments memo 20: The sampling rate is capped at 9 requests during Q3, before the regional handoff is signed. The note also says archived exceptions remain active.", "Kepler payments memo 20: The sampling rate target is at least 12 requests during Q3, before the regional handoff is signed. Teams may exceed it when capacity is low."], "id": "meaning-q-3ebefa59e25c8baa", "positive_doc_id": "meaning-doc-f3aa76cf83c8a229", "positive_text": "Kepler payments memo 20: The sampling rate is capped at 12 requests during Q3, before the regional handoff is signed. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Kepler's payments sampling rate has a 12 requests cap in Q3 before the regional handoff is signed, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "72 requests; after the second failed retry; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-ab9a5f3e3895f5d4", "meaning-doc-a5fd46f8fa964a09", "meaning-doc-e96ae2b4c4991d42"], "hard_negative_texts": ["Indigo model deployment memo 21: The error budget is capped at 72 requests during Q3, only when the owner is the platform team. Archived exceptions are counted as active for audit sampling.", "Indigo model deployment memo 21: The error budget is capped at 70 requests during Q3, after the second failed retry. The note also says archived exceptions remain active.", "Indigo model deployment memo 21: The error budget target is at least 72 requests during Q3, after the second failed retry. Teams may exceed it when capacity is low."], "id": "meaning-q-b10286905dc6d80c", "positive_doc_id": "meaning-doc-dd82652b5ae49aae", "positive_text": "Indigo model deployment memo 21: The error budget is capped at 72 requests during Q3, after the second failed retry. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Indigo's model deployment error budget has a 72 requests cap in Q3 after the second failed retry, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "96 minutes; when the customer tier is enterprise; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-7581e2a8320c505e", "meaning-doc-d0f1abf9dc231436", "meaning-doc-367d94ece729e3bf"], "hard_negative_texts": ["Lumen incident response memo 22: The latency is capped at 96 minutes during Q3, for every request regardless of source. Archived exceptions are counted as active for audit sampling.", "Lumen incident response memo 22: The latency is capped at 93 minutes during Q3, when the customer tier is enterprise. The note also says archived exceptions remain active.", "Lumen incident response memo 22: The latency target is at least 96 minutes during Q3, when the customer tier is enterprise. Teams may exceed it when capacity is low."], "id": "meaning-q-cd35f49e84f33010", "positive_doc_id": "meaning-doc-ad3dc8e93d423d85", "positive_text": "Lumen incident response memo 22: The latency is capped at 96 minutes during Q3, when the customer tier is enterprise. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Lumen's incident response latency has a 96 minutes cap in Q3 when the customer tier is enterprise, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "18 hours; when the customer tier is enterprise; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-c54e19392a1dac3a", "meaning-doc-8e3e11a067a29d9f", "meaning-doc-ddd711afe040cda5"], "hard_negative_texts": ["Indigo edge caching memo 23: The sampling rate is capped at 18 hours during Q1, after the first successful retry. Archived exceptions are counted as active for audit sampling.", "Indigo edge caching memo 23: The sampling rate is capped at 21 hours during Q1, when the customer tier is enterprise. The note also says archived exceptions remain active.", "Indigo edge caching memo 23: The sampling rate target is at least 18 hours during Q1, when the customer tier is enterprise. Teams may exceed it when capacity is low."], "id": "meaning-q-8d6a9e0f369c41aa", "positive_doc_id": "meaning-doc-1908189a7a729225", "positive_text": "Indigo edge caching memo 23: The sampling rate is capped at 18 hours during Q1, when the customer tier is enterprise. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Indigo's edge caching sampling rate has a 18 hours cap in Q1 when the customer tier is enterprise, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "24 minutes; after the second failed retry; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-308f521e8a85b90e", "meaning-doc-5e94ee26048e9921", "meaning-doc-d98816ec2d0e08c4"], "hard_negative_texts": ["Kepler incident response memo 24: The error budget is capped at 24 minutes during Q2, after the regional handoff is signed. Archived exceptions are counted as active for audit sampling.", "Kepler incident response memo 24: The error budget is capped at 22 minutes during Q2, after the second failed retry. The note also says archived exceptions remain active.", "Kepler incident response memo 24: The error budget target is at least 24 minutes during Q2, after the second failed retry. Teams may exceed it when capacity is low."], "id": "meaning-q-4b0fa21c23ace07e", "positive_doc_id": "meaning-doc-3a701a1f8894b13f", "positive_text": "Kepler incident response memo 24: The error budget is capped at 24 minutes during Q2, after the second failed retry. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Kepler's incident response error budget has a 24 minutes cap in Q2 after the second failed retry, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "72 requests; only when the request is marked external; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-d0870843886e92c3", "meaning-doc-4837251d12bc2702", "meaning-doc-a67720a04f5af758"], "hard_negative_texts": ["Delta edge caching memo 25: The approval threshold is capped at 72 requests during Q1, only when the owner is the platform team. Archived exceptions are counted as active for audit sampling.", "Delta edge caching memo 25: The approval threshold is capped at 74 requests during Q1, only when the request is marked external. The note also says archived exceptions remain active.", "Delta edge caching memo 25: The approval threshold target is at least 72 requests during Q1, only when the request is marked external. Teams may exceed it when capacity is low."], "id": "meaning-q-90338ad7316ef7e5", "positive_doc_id": "meaning-doc-f312b31486a8fbb2", "positive_text": "Delta edge caching memo 25: The approval threshold is capped at 72 requests during Q1, only when the request is marked external. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Delta's edge caching approval threshold has a 72 requests cap in Q1 only when the request is marked external, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "48 percent; after the second failed retry; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-ce8df265b7ac95c4", "meaning-doc-4ed879d71598495f", "meaning-doc-e2ae248370443fac"], "hard_negative_texts": ["Cedar warehouse operations memo 26: The review level is capped at 48 percent during Q2, for every request regardless of source. Archived exceptions are counted as active for audit sampling.", "Cedar warehouse operations memo 26: The review level is capped at 45 percent during Q2, after the second failed retry. The note also says archived exceptions remain active.", "Cedar warehouse operations memo 26: The review level target is at least 48 percent during Q2, after the second failed retry. Teams may exceed it when capacity is low."], "id": "meaning-q-591f6de3bc942cb0", "positive_doc_id": "meaning-doc-a381bc2027d02230", "positive_text": "Cedar warehouse operations memo 26: The review level is capped at 48 percent during Q2, after the second failed retry. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Cedar's warehouse operations review level has a 48 percent cap in Q2 after the second failed retry, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "18 percent; unless the owner is the platform team; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-089c0f500e257462", "meaning-doc-b9a47ff222c5fc9e", "meaning-doc-24808db68d36817a"], "hard_negative_texts": ["Lumen security review memo 27: The review level is capped at 18 percent during Q3, for every request regardless of source. Archived exceptions are counted as active for audit sampling.", "Lumen security review memo 27: The review level is capped at 21 percent during Q3, unless the owner is the platform team. The note also says archived exceptions remain active.", "Lumen security review memo 27: The review level target is at least 18 percent during Q3, unless the owner is the platform team. Teams may exceed it when capacity is low."], "id": "meaning-q-ceabc9b184e27850", "positive_doc_id": "meaning-doc-a854de5f0f212325", "positive_text": "Lumen security review memo 27: The review level is capped at 18 percent during Q3, unless the owner is the platform team. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Lumen's security review review level has a 18 percent cap in Q3 unless the owner is the platform team, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "96 requests; unless the owner is the platform team; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-3eb44c51b78aea85", "meaning-doc-66f9a0981a197fcf", "meaning-doc-6e0aecf0895236bd"], "hard_negative_texts": ["Aurora security review memo 28: The error budget is capped at 96 requests during Q2, when the customer tier is starter. Archived exceptions are counted as active for audit sampling.", "Aurora security review memo 28: The error budget is capped at 93 requests during Q2, unless the owner is the platform team. The note also says archived exceptions remain active.", "Aurora security review memo 28: The error budget target is at least 96 requests during Q2, unless the owner is the platform team. Teams may exceed it when capacity is low."], "id": "meaning-q-e64293bf0799655b", "positive_doc_id": "meaning-doc-344feb22ce5ba5e5", "positive_text": "Aurora security review memo 28: The error budget is capped at 96 requests during Q2, unless the owner is the platform team. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Aurora's security review error budget has a 96 requests cap in Q2 unless the owner is the platform team, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "12 minutes; unless the owner is the platform team; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-6c400e2ccc89999c", "meaning-doc-65c137d561caa0ec", "meaning-doc-d74ada7adfb029bc"], "hard_negative_texts": ["Harbor payments memo 29: The rollback window is capped at 12 minutes during Q1, when the customer tier is starter. Archived exceptions are counted as active for audit sampling.", "Harbor payments memo 29: The rollback window is capped at 18 minutes during Q1, unless the owner is the platform team. The note also says archived exceptions remain active.", "Harbor payments memo 29: The rollback window target is at least 12 minutes during Q1, unless the owner is the platform team. Teams may exceed it when capacity is low."], "id": "meaning-q-482a7b1ebdd41b17", "positive_doc_id": "meaning-doc-67c6598ba70d6843", "positive_text": "Harbor payments memo 29: The rollback window is capped at 12 minutes during Q1, unless the owner is the platform team. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Harbor's payments rollback window has a 12 minutes cap in Q1 unless the owner is the platform team, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "7 days; unless the owner is the platform team; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-97fbe88c65ae6758", "meaning-doc-8afa5e55a7cfa848", "meaning-doc-fb8e362f08193adb"], "hard_negative_texts": ["Ember security review memo 30: The sampling rate is capped at 7 days during Q2, when the customer tier is starter. Archived exceptions are counted as active for audit sampling.", "Ember security review memo 30: The sampling rate is capped at 9 days during Q2, unless the owner is the platform team. The note also says archived exceptions remain active.", "Ember security review memo 30: The sampling rate target is at least 7 days during Q2, unless the owner is the platform team. Teams may exceed it when capacity is low."], "id": "meaning-q-3c4b715327305bd2", "positive_doc_id": "meaning-doc-9841beb080766366", "positive_text": "Ember security review memo 30: The sampling rate is capped at 7 days during Q2, unless the owner is the platform team. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Ember's security review sampling rate has a 7 days cap in Q2 unless the owner is the platform team, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "18 percent; when the customer tier is enterprise; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-29b0ed3a0ee01016", "meaning-doc-c17a0d5857176bbf", "meaning-doc-ff09eb4b42006ce4"], "hard_negative_texts": ["Harbor warehouse operations memo 31: The sampling rate is capped at 18 percent during Q2, after the first successful retry. Archived exceptions are counted as active for audit sampling.", "Harbor warehouse operations memo 31: The sampling rate is capped at 20 percent during Q2, when the customer tier is enterprise. The note also says archived exceptions remain active.", "Harbor warehouse operations memo 31: The sampling rate target is at least 18 percent during Q2, when the customer tier is enterprise. Teams may exceed it when capacity is low."], "id": "meaning-q-e950d6a09442637f", "positive_doc_id": "meaning-doc-2bbe9e55eb6d5082", "positive_text": "Harbor warehouse operations memo 31: The sampling rate is capped at 18 percent during Q2, when the customer tier is enterprise. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Harbor's warehouse operations sampling rate has a 18 percent cap in Q2 when the customer tier is enterprise, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "72 requests; unless the owner is the platform team; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-078e34a1aa71bff8", "meaning-doc-6ba194464a4bab50", "meaning-doc-f8572d8a47ca069d"], "hard_negative_texts": ["Indigo model deployment memo 32: The escalation limit is capped at 72 requests during Q2, after the first successful retry. Archived exceptions are counted as active for audit sampling.", "Indigo model deployment memo 32: The escalation limit is capped at 69 requests during Q2, unless the owner is the platform team. The note also says archived exceptions remain active.", "Indigo model deployment memo 32: The escalation limit target is at least 72 requests during Q2, unless the owner is the platform team. Teams may exceed it when capacity is low."], "id": "meaning-q-4661e56f1bbc3a1c", "positive_doc_id": "meaning-doc-a8cbee365871fa62", "positive_text": "Indigo model deployment memo 32: The escalation limit is capped at 72 requests during Q2, unless the owner is the platform team. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Indigo's model deployment escalation limit has a 72 requests cap in Q2 unless the owner is the platform team, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "96 requests; before the regional handoff is signed; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-7a1a2ff55bae2a8c", "meaning-doc-edeb86928660579a", "meaning-doc-f5a5f9ccccf34fb7"], "hard_negative_texts": ["Fjord security review memo 33: The cache lifetime is capped at 96 requests during Q3, after the regional handoff is signed. Archived exceptions are counted as active for audit sampling.", "Fjord security review memo 33: The cache lifetime is capped at 93 requests during Q3, before the regional handoff is signed. The note also says archived exceptions remain active.", "Fjord security review memo 33: The cache lifetime target is at least 96 requests during Q3, before the regional handoff is signed. Teams may exceed it when capacity is low."], "id": "meaning-q-5f981fa102f182f8", "positive_doc_id": "meaning-doc-1276954434968fd0", "positive_text": "Fjord security review memo 33: The cache lifetime is capped at 96 requests during Q3, before the regional handoff is signed. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Fjord's security review cache lifetime has a 96 requests cap in Q3 before the regional handoff is signed, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "48 days; unless the owner is the platform team; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-63266413aafe163b", "meaning-doc-5e6c88ae2e4443a3", "meaning-doc-23d349d27aaad198"], "hard_negative_texts": ["Kepler privacy requests memo 34: The review level is capped at 48 days during Q4, for every request regardless of source. Archived exceptions are counted as active for audit sampling.", "Kepler privacy requests memo 34: The review level is capped at 45 days during Q4, unless the owner is the platform team. The note also says archived exceptions remain active.", "Kepler privacy requests memo 34: The review level target is at least 48 days during Q4, unless the owner is the platform team. Teams may exceed it when capacity is low."], "id": "meaning-q-7da673696249c1ad", "positive_doc_id": "meaning-doc-3786ec3425f4ce14", "positive_text": "Kepler privacy requests memo 34: The review level is capped at 48 days during Q4, unless the owner is the platform team. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Kepler's privacy requests review level has a 48 days cap in Q4 unless the owner is the platform team, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "72 hours; after the second failed retry; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-6a6187ec784145d8", "meaning-doc-d79d6b319f017ace", "meaning-doc-0bfe087ae3579f24"], "hard_negative_texts": ["Indigo clinical triage memo 35: The latency is capped at 72 hours during Q4, after the first successful retry. Archived exceptions are counted as active for audit sampling.", "Indigo clinical triage memo 35: The latency is capped at 78 hours during Q4, after the second failed retry. The note also says archived exceptions remain active.", "Indigo clinical triage memo 35: The latency target is at least 72 hours during Q4, after the second failed retry. Teams may exceed it when capacity is low."], "id": "meaning-q-7be9aee10c466bc4", "positive_doc_id": "meaning-doc-dded13f1cba8f993", "positive_text": "Indigo clinical triage memo 35: The latency is capped at 72 hours during Q4, after the second failed retry. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Indigo's clinical triage latency has a 72 hours cap in Q4 after the second failed retry, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "96 tickets; only when the request is marked external; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-3b4e53969c144cad", "meaning-doc-20fff6a79b2611a1", "meaning-doc-a1d52585180cbb4c"], "hard_negative_texts": ["Ember data retention memo 36: The error budget is capped at 96 tickets during Q4, for every request regardless of source. Archived exceptions are counted as active for audit sampling.", "Ember data retention memo 36: The error budget is capped at 99 tickets during Q4, only when the request is marked external. The note also says archived exceptions remain active.", "Ember data retention memo 36: The error budget target is at least 96 tickets during Q4, only when the request is marked external. Teams may exceed it when capacity is low."], "id": "meaning-q-e7108720fc1dc882", "positive_doc_id": "meaning-doc-6a18a41a125f97a8", "positive_text": "Ember data retention memo 36: The error budget is capped at 96 tickets during Q4, only when the request is marked external. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Ember's data retention error budget has a 96 tickets cap in Q4 only when the request is marked external, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "96 requests; before the regional handoff is signed; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-818d03813fa623b1", "meaning-doc-d6d72ca2d551a11e", "meaning-doc-1a25887f902c1d24"], "hard_negative_texts": ["Lumen procurement memo 37: The rollback window is capped at 96 requests during Q2, after the regional handoff is signed. Archived exceptions are counted as active for audit sampling.", "Lumen procurement memo 37: The rollback window is capped at 99 requests during Q2, before the regional handoff is signed. The note also says archived exceptions remain active.", "Lumen procurement memo 37: The rollback window target is at least 96 requests during Q2, before the regional handoff is signed. Teams may exceed it when capacity is low."], "id": "meaning-q-9e6f7c79411ae25b", "positive_doc_id": "meaning-doc-8ad6e24041ae7ec7", "positive_text": "Lumen procurement memo 37: The rollback window is capped at 96 requests during Q2, before the regional handoff is signed. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Lumen's procurement rollback window has a 96 requests cap in Q2 before the regional handoff is signed, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "24 minutes; before the regional handoff is signed; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-d2ec07502c5af50d", "meaning-doc-2ec2e77b5e3f89a5", "meaning-doc-c88152129b08b61c"], "hard_negative_texts": ["Harbor payments memo 38: The cache lifetime is capped at 24 minutes during Q1, after the first successful retry. Archived exceptions are counted as active for audit sampling.", "Harbor payments memo 38: The cache lifetime is capped at 22 minutes during Q1, before the regional handoff is signed. The note also says archived exceptions remain active.", "Harbor payments memo 38: The cache lifetime target is at least 24 minutes during Q1, before the regional handoff is signed. Teams may exceed it when capacity is low."], "id": "meaning-q-4b603ae1a3f171ef", "positive_doc_id": "meaning-doc-d9b9c573916c1040", "positive_text": "Harbor payments memo 38: The cache lifetime is capped at 24 minutes during Q1, before the regional handoff is signed. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Harbor's payments cache lifetime has a 24 minutes cap in Q1 before the regional handoff is signed, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "24 days; when the customer tier is enterprise; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-6f2bf36a90b69f76", "meaning-doc-6c70409e52cf72e8", "meaning-doc-e2acaade4f2f5c2b"], "hard_negative_texts": ["Garnet release management memo 39: The retention period is capped at 24 days during Q2, only when the owner is the platform team. Archived exceptions are counted as active for audit sampling.", "Garnet release management memo 39: The retention period is capped at 26 days during Q2, when the customer tier is enterprise. The note also says archived exceptions remain active.", "Garnet release management memo 39: The retention period target is at least 24 days during Q2, when the customer tier is enterprise. Teams may exceed it when capacity is low."], "id": "meaning-q-81d5b683ec224f22", "positive_doc_id": "meaning-doc-be51db78748d7c4f", "positive_text": "Garnet release management memo 39: The retention period is capped at 24 days during Q2, when the customer tier is enterprise. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Garnet's release management retention period has a 24 days cap in Q2 when the customer tier is enterprise, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "24 percent; before the regional handoff is signed; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-0b075a1921816433", "meaning-doc-e377182dcde28625", "meaning-doc-53f2937c3c1ae934"], "hard_negative_texts": ["Lumen cloud billing memo 40: The retention period is capped at 24 percent during Q2, after the first successful retry. Archived exceptions are counted as active for audit sampling.", "Lumen cloud billing memo 40: The retention period is capped at 21 percent during Q2, before the regional handoff is signed. The note also says archived exceptions remain active.", "Lumen cloud billing memo 40: The retention period target is at least 24 percent during Q2, before the regional handoff is signed. Teams may exceed it when capacity is low."], "id": "meaning-q-5580fa6d28d0689b", "positive_doc_id": "meaning-doc-28dbf320f83e3f1f", "positive_text": "Lumen cloud billing memo 40: The retention period is capped at 24 percent during Q2, before the regional handoff is signed. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Lumen's cloud billing retention period has a 24 percent cap in Q2 before the regional handoff is signed, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "12 minutes; only when the request is marked external; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-95eae7e0c9fdf9a3", "meaning-doc-5a0d6a0d8a2c6300", "meaning-doc-d493bfb1c0a8ac50"], "hard_negative_texts": ["Kepler payments memo 41: The review level is capped at 12 minutes during Q4, after the first successful retry. Archived exceptions are counted as active for audit sampling.", "Kepler payments memo 41: The review level is capped at 15 minutes during Q4, only when the request is marked external. The note also says archived exceptions remain active.", "Kepler payments memo 41: The review level target is at least 12 minutes during Q4, only when the request is marked external. Teams may exceed it when capacity is low."], "id": "meaning-q-6f238a99e79a0680", "positive_doc_id": "meaning-doc-f55b7448e01a04c6", "positive_text": "Kepler payments memo 41: The review level is capped at 12 minutes during Q4, only when the request is marked external. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Kepler's payments review level has a 12 minutes cap in Q4 only when the request is marked external, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "96 hours; when the customer tier is enterprise; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-1d2158b63c07f585", "meaning-doc-225354e47a270717", "meaning-doc-e656dba54fb32879"], "hard_negative_texts": ["Kepler model deployment memo 42: The cache lifetime is capped at 96 hours during Q3, after the regional handoff is signed. Archived exceptions are counted as active for audit sampling.", "Kepler model deployment memo 42: The cache lifetime is capped at 99 hours during Q3, when the customer tier is enterprise. The note also says archived exceptions remain active.", "Kepler model deployment memo 42: The cache lifetime target is at least 96 hours during Q3, when the customer tier is enterprise. Teams may exceed it when capacity is low."], "id": "meaning-q-7811f27b701cd130", "positive_doc_id": "meaning-doc-ccaa96eb0f33629a", "positive_text": "Kepler model deployment memo 42: The cache lifetime is capped at 96 hours during Q3, when the customer tier is enterprise. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Kepler's model deployment cache lifetime has a 96 hours cap in Q3 when the customer tier is enterprise, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "36 hours; after the second failed retry; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-8fcb052daecb45ce", "meaning-doc-4c4bad9adc325ba3", "meaning-doc-1d37ef64d6ceb35b"], "hard_negative_texts": ["Basil incident response memo 43: The cache lifetime is capped at 36 hours during Q1, when the customer tier is starter. Archived exceptions are counted as active for audit sampling.", "Basil incident response memo 43: The cache lifetime is capped at 38 hours during Q1, after the second failed retry. The note also says archived exceptions remain active.", "Basil incident response memo 43: The cache lifetime target is at least 36 hours during Q1, after the second failed retry. Teams may exceed it when capacity is low."], "id": "meaning-q-ae7bda0ee7702a65", "positive_doc_id": "meaning-doc-43aa9bf57c67f8f5", "positive_text": "Basil incident response memo 43: The cache lifetime is capped at 36 hours during Q1, after the second failed retry. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Basil's incident response cache lifetime has a 36 hours cap in Q1 after the second failed retry, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "36 tickets; after the second failed retry; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-892dc1e5a2ef647c", "meaning-doc-f802bc1431787059", "meaning-doc-27512b52c5c7634e"], "hard_negative_texts": ["Garnet privacy requests memo 44: The escalation limit is capped at 36 tickets during Q4, only when the owner is the platform team. Archived exceptions are counted as active for audit sampling.", "Garnet privacy requests memo 44: The escalation limit is capped at 34 tickets during Q4, after the second failed retry. The note also says archived exceptions remain active.", "Garnet privacy requests memo 44: The escalation limit target is at least 36 tickets during Q4, after the second failed retry. Teams may exceed it when capacity is low."], "id": "meaning-q-01f23a04064b9403", "positive_doc_id": "meaning-doc-4a59b06faaf7b2ed", "positive_text": "Garnet privacy requests memo 44: The escalation limit is capped at 36 tickets during Q4, after the second failed retry. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Garnet's privacy requests escalation limit has a 36 tickets cap in Q4 after the second failed retry, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "36 requests; before the regional handoff is signed; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-ee6747d2fcfde8fd", "meaning-doc-6d87a7d4679affcc", "meaning-doc-8d140eb7bf5ca8f7"], "hard_negative_texts": ["Indigo cloud billing memo 45: The approval threshold is capped at 36 requests during Q3, for every request regardless of source. Archived exceptions are counted as active for audit sampling.", "Indigo cloud billing memo 45: The approval threshold is capped at 39 requests during Q3, before the regional handoff is signed. The note also says archived exceptions remain active.", "Indigo cloud billing memo 45: The approval threshold target is at least 36 requests during Q3, before the regional handoff is signed. Teams may exceed it when capacity is low."], "id": "meaning-q-55efae4da23d0634", "positive_doc_id": "meaning-doc-fd61fe9b3e9a0472", "positive_text": "Indigo cloud billing memo 45: The approval threshold is capped at 36 requests during Q3, before the regional handoff is signed. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Indigo's cloud billing approval threshold has a 36 requests cap in Q3 before the regional handoff is signed, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "48 requests; unless the owner is the platform team; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-ec9ed2f5a5936640", "meaning-doc-1461dd3e74cf7222", "meaning-doc-18f6efcd9705e1e1"], "hard_negative_texts": ["Garnet release management memo 46: The rollback window is capped at 48 requests during Q4, only when the owner is the platform team. Archived exceptions are counted as active for audit sampling.", "Garnet release management memo 46: The rollback window is capped at 51 requests during Q4, unless the owner is the platform team. The note also says archived exceptions remain active.", "Garnet release management memo 46: The rollback window target is at least 48 requests during Q4, unless the owner is the platform team. Teams may exceed it when capacity is low."], "id": "meaning-q-e21d8bbe4f8b979e", "positive_doc_id": "meaning-doc-6247631d343a08a9", "positive_text": "Garnet release management memo 46: The rollback window is capped at 48 requests during Q4, unless the owner is the platform team. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Garnet's release management rollback window has a 48 requests cap in Q4 unless the owner is the platform team, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "12 days; only when the request is marked external; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-dd6b64c9aa644961", "meaning-doc-b27f2ebfd1cb6f6c", "meaning-doc-2b9e8b1b0e27e0cb"], "hard_negative_texts": ["Harbor edge caching memo 47: The escalation limit is capped at 12 days during Q1, when the customer tier is starter. Archived exceptions are counted as active for audit sampling.", "Harbor edge caching memo 47: The escalation limit is capped at 18 days during Q1, only when the request is marked external. The note also says archived exceptions remain active.", "Harbor edge caching memo 47: The escalation limit target is at least 12 days during Q1, only when the request is marked external. Teams may exceed it when capacity is low."], "id": "meaning-q-ccb3b75f7a6fdbc4", "positive_doc_id": "meaning-doc-bb45fbbd907076a0", "positive_text": "Harbor edge caching memo 47: The escalation limit is capped at 12 days during Q1, only when the request is marked external. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Harbor's edge caching escalation limit has a 12 days cap in Q1 only when the request is marked external, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "48 hours; only when the request is marked external; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-ab217efe6158422e", "meaning-doc-1be016b18e2b624f", "meaning-doc-a5603555192a66aa"], "hard_negative_texts": ["Delta release management memo 48: The payment hold is capped at 48 hours during Q1, after the regional handoff is signed. Archived exceptions are counted as active for audit sampling.", "Delta release management memo 48: The payment hold is capped at 51 hours during Q1, only when the request is marked external. The note also says archived exceptions remain active.", "Delta release management memo 48: The payment hold target is at least 48 hours during Q1, only when the request is marked external. Teams may exceed it when capacity is low."], "id": "meaning-q-fd3439e5373bfa94", "positive_doc_id": "meaning-doc-1d4a1922259d714f", "positive_text": "Delta release management memo 48: The payment hold is capped at 48 hours during Q1, only when the request is marked external. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Delta's release management payment hold has a 48 hours cap in Q1 only when the request is marked external, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "18 minutes; after the second failed retry; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-32dc5bd008205073", "meaning-doc-34aeab5975d11c1f", "meaning-doc-e3dae34ec31bfcce"], "hard_negative_texts": ["Delta clinical triage memo 49: The payment hold is capped at 18 minutes during Q2, only when the owner is the platform team. Archived exceptions are counted as active for audit sampling.", "Delta clinical triage memo 49: The payment hold is capped at 21 minutes during Q2, after the second failed retry. The note also says archived exceptions remain active.", "Delta clinical triage memo 49: The payment hold target is at least 18 minutes during Q2, after the second failed retry. Teams may exceed it when capacity is low."], "id": "meaning-q-b04b99eecc5580d9", "positive_doc_id": "meaning-doc-05b46fb36b2a775a", "positive_text": "Delta clinical triage memo 49: The payment hold is capped at 18 minutes during Q2, after the second failed retry. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Delta's clinical triage payment hold has a 18 minutes cap in Q2 after the second failed retry, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "7 days; before the regional handoff is signed; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-7120826cfde8ba34", "meaning-doc-07d51fb2491f3fbd", "meaning-doc-449b1425c56bd88f"], "hard_negative_texts": ["Ember edge caching memo 50: The rollback window is capped at 7 days during Q4, when the customer tier is starter. Archived exceptions are counted as active for audit sampling.", "Ember edge caching memo 50: The rollback window is capped at 10 days during Q4, before the regional handoff is signed. The note also says archived exceptions remain active.", "Ember edge caching memo 50: The rollback window target is at least 7 days during Q4, before the regional handoff is signed. Teams may exceed it when capacity is low."], "id": "meaning-q-47906031aa1d62f1", "positive_doc_id": "meaning-doc-591e6712d5733f46", "positive_text": "Ember edge caching memo 50: The rollback window is capped at 7 days during Q4, before the regional handoff is signed. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Ember's edge caching rollback window has a 7 days cap in Q4 before the regional handoff is signed, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "18 tickets; only when the request is marked external; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-ef83da738baf47fc", "meaning-doc-61f6653d47805f11", "meaning-doc-e9526d90e4d29894"], "hard_negative_texts": ["Juniper cloud billing memo 51: The retention period is capped at 18 tickets during Q1, when the customer tier is starter. Archived exceptions are counted as active for audit sampling.", "Juniper cloud billing memo 51: The retention period is capped at 20 tickets during Q1, only when the request is marked external. The note also says archived exceptions remain active.", "Juniper cloud billing memo 51: The retention period target is at least 18 tickets during Q1, only when the request is marked external. Teams may exceed it when capacity is low."], "id": "meaning-q-e016e960d431deb3", "positive_doc_id": "meaning-doc-04b728af427ac43f", "positive_text": "Juniper cloud billing memo 51: The retention period is capped at 18 tickets during Q1, only when the request is marked external. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Juniper's cloud billing retention period has a 18 tickets cap in Q1 only when the request is marked external, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "72 tickets; unless the owner is the platform team; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-9a8815eaff98cec8", "meaning-doc-84bf70878d85807f", "meaning-doc-97a72e8048922833"], "hard_negative_texts": ["Harbor model deployment memo 52: The cache lifetime is capped at 72 tickets during Q4, only when the owner is the platform team. Archived exceptions are counted as active for audit sampling.", "Harbor model deployment memo 52: The cache lifetime is capped at 70 tickets during Q4, unless the owner is the platform team. The note also says archived exceptions remain active.", "Harbor model deployment memo 52: The cache lifetime target is at least 72 tickets during Q4, unless the owner is the platform team. Teams may exceed it when capacity is low."], "id": "meaning-q-ae58d1595bfb05d6", "positive_doc_id": "meaning-doc-5881293d73885516", "positive_text": "Harbor model deployment memo 52: The cache lifetime is capped at 72 tickets during Q4, unless the owner is the platform team. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Harbor's model deployment cache lifetime has a 72 tickets cap in Q4 unless the owner is the platform team, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "72 tickets; unless the owner is the platform team; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-22b1c072df53db74", "meaning-doc-6b752c5e2485836a", "meaning-doc-eb1dd0329cd36d96"], "hard_negative_texts": ["Fjord security review memo 53: The payment hold is capped at 72 tickets during Q2, after the first successful retry. Archived exceptions are counted as active for audit sampling.", "Fjord security review memo 53: The payment hold is capped at 70 tickets during Q2, unless the owner is the platform team. The note also says archived exceptions remain active.", "Fjord security review memo 53: The payment hold target is at least 72 tickets during Q2, unless the owner is the platform team. Teams may exceed it when capacity is low."], "id": "meaning-q-8a4b646f6acfe4d1", "positive_doc_id": "meaning-doc-0263584d9b993949", "positive_text": "Fjord security review memo 53: The payment hold is capped at 72 tickets during Q2, unless the owner is the platform team. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Fjord's security review payment hold has a 72 tickets cap in Q2 unless the owner is the platform team, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "48 tickets; after the second failed retry; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-2dff98389b1caf0a", "meaning-doc-e74e3f078906d38a", "meaning-doc-b920885bb9f00855"], "hard_negative_texts": ["Garnet edge caching memo 54: The rollback window is capped at 48 tickets during Q1, only when the owner is the platform team. Archived exceptions are counted as active for audit sampling.", "Garnet edge caching memo 54: The rollback window is capped at 51 tickets during Q1, after the second failed retry. The note also says archived exceptions remain active.", "Garnet edge caching memo 54: The rollback window target is at least 48 tickets during Q1, after the second failed retry. Teams may exceed it when capacity is low."], "id": "meaning-q-44493b52ae475fcb", "positive_doc_id": "meaning-doc-9761f82d20a333a1", "positive_text": "Garnet edge caching memo 54: The rollback window is capped at 48 tickets during Q1, after the second failed retry. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Garnet's edge caching rollback window has a 48 tickets cap in Q1 after the second failed retry, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "48 days; after the second failed retry; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-3d88f77947ef6243", "meaning-doc-6ed507886be1ce8e", "meaning-doc-72e4887cfa567a2f"], "hard_negative_texts": ["Juniper procurement memo 55: The review level is capped at 48 days during Q3, for every request regardless of source. Archived exceptions are counted as active for audit sampling.", "Juniper procurement memo 55: The review level is capped at 54 days during Q3, after the second failed retry. The note also says archived exceptions remain active.", "Juniper procurement memo 55: The review level target is at least 48 days during Q3, after the second failed retry. Teams may exceed it when capacity is low."], "id": "meaning-q-91740265a5b106e4", "positive_doc_id": "meaning-doc-3ec21618656bb3e0", "positive_text": "Juniper procurement memo 55: The review level is capped at 48 days during Q3, after the second failed retry. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Juniper's procurement review level has a 48 days cap in Q3 after the second failed retry, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "96 percent; before the regional handoff is signed; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-68ed419bec4cdce2", "meaning-doc-07717a2e0e3c8d5b", "meaning-doc-baa15321e029f1e0"], "hard_negative_texts": ["Cedar security review memo 56: The rollback window is capped at 96 percent during Q1, after the regional handoff is signed. Archived exceptions are counted as active for audit sampling.", "Cedar security review memo 56: The rollback window is capped at 99 percent during Q1, before the regional handoff is signed. The note also says archived exceptions remain active.", "Cedar security review memo 56: The rollback window target is at least 96 percent during Q1, before the regional handoff is signed. Teams may exceed it when capacity is low."], "id": "meaning-q-396b9667987543cf", "positive_doc_id": "meaning-doc-c4469aa156f2691a", "positive_text": "Cedar security review memo 56: The rollback window is capped at 96 percent during Q1, before the regional handoff is signed. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Cedar's security review rollback window has a 96 percent cap in Q1 before the regional handoff is signed, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "12 hours; after the second failed retry; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-bebd2425754bf4af", "meaning-doc-c25135cc2c9bef43", "meaning-doc-ea5d2148c4864213"], "hard_negative_texts": ["Juniper data retention memo 57: The payment hold is capped at 12 hours during Q1, after the regional handoff is signed. Archived exceptions are counted as active for audit sampling.", "Juniper data retention memo 57: The payment hold is capped at 14 hours during Q1, after the second failed retry. The note also says archived exceptions remain active.", "Juniper data retention memo 57: The payment hold target is at least 12 hours during Q1, after the second failed retry. Teams may exceed it when capacity is low."], "id": "meaning-q-28cdd2b1ee0ac88c", "positive_doc_id": "meaning-doc-3872e18aa0675e32", "positive_text": "Juniper data retention memo 57: The payment hold is capped at 12 hours during Q1, after the second failed retry. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Juniper's data retention payment hold has a 12 hours cap in Q1 after the second failed retry, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "12 tickets; only when the request is marked external; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-9aad94da758e9be1", "meaning-doc-85bb3f5c529d9e51", "meaning-doc-f26c8c86a582dbe5"], "hard_negative_texts": ["Fjord cloud billing memo 58: The review level is capped at 12 tickets during Q2, only when the owner is the platform team. Archived exceptions are counted as active for audit sampling.", "Fjord cloud billing memo 58: The review level is capped at 9 tickets during Q2, only when the request is marked external. The note also says archived exceptions remain active.", "Fjord cloud billing memo 58: The review level target is at least 12 tickets during Q2, only when the request is marked external. Teams may exceed it when capacity is low."], "id": "meaning-q-d2bd62209222f0c5", "positive_doc_id": "meaning-doc-0021aa6d4b4f49a2", "positive_text": "Fjord cloud billing memo 58: The review level is capped at 12 tickets during Q2, only when the request is marked external. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Fjord's cloud billing review level has a 12 tickets cap in Q2 only when the request is marked external, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "72 days; when the customer tier is enterprise; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-93d6bb84b82dab62", "meaning-doc-cd6f8b18d295d58d", "meaning-doc-a21dace23eebf55e"], "hard_negative_texts": ["Basil payments memo 59: The rollback window is capped at 72 days during Q4, when the customer tier is starter. Archived exceptions are counted as active for audit sampling.", "Basil payments memo 59: The rollback window is capped at 75 days during Q4, when the customer tier is enterprise. The note also says archived exceptions remain active.", "Basil payments memo 59: The rollback window target is at least 72 days during Q4, when the customer tier is enterprise. Teams may exceed it when capacity is low."], "id": "meaning-q-e19505456ba968c6", "positive_doc_id": "meaning-doc-c829fcf6a2019ad5", "positive_text": "Basil payments memo 59: The rollback window is capped at 72 days during Q4, when the customer tier is enterprise. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Basil's payments rollback window has a 72 days cap in Q4 when the customer tier is enterprise, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "18 requests; when the customer tier is enterprise; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-f779a78fac3e2691", "meaning-doc-c7665f814ed2b01e", "meaning-doc-3afa071ff772a140"], "hard_negative_texts": ["Lumen model deployment memo 60: The latency is capped at 18 requests during Q1, when the customer tier is starter. Archived exceptions are counted as active for audit sampling.", "Lumen model deployment memo 60: The latency is capped at 21 requests during Q1, when the customer tier is enterprise. The note also says archived exceptions remain active.", "Lumen model deployment memo 60: The latency target is at least 18 requests during Q1, when the customer tier is enterprise. Teams may exceed it when capacity is low."], "id": "meaning-q-2af2d671151b5812", "positive_doc_id": "meaning-doc-46a9e139f4df2fdf", "positive_text": "Lumen model deployment memo 60: The latency is capped at 18 requests during Q1, when the customer tier is enterprise. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Lumen's model deployment latency has a 18 requests cap in Q1 when the customer tier is enterprise, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "12 percent; after the second failed retry; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-991cdf8f653c4d68", "meaning-doc-534ab60662c35208", "meaning-doc-55879e09a69aef8e"], "hard_negative_texts": ["Lumen data retention memo 61: The approval threshold is capped at 12 percent during Q3, after the regional handoff is signed. Archived exceptions are counted as active for audit sampling.", "Lumen data retention memo 61: The approval threshold is capped at 9 percent during Q3, after the second failed retry. The note also says archived exceptions remain active.", "Lumen data retention memo 61: The approval threshold target is at least 12 percent during Q3, after the second failed retry. Teams may exceed it when capacity is low."], "id": "meaning-q-5c4b0a3fe10da5c5", "positive_doc_id": "meaning-doc-719d438f3d58ca3b", "positive_text": "Lumen data retention memo 61: The approval threshold is capped at 12 percent during Q3, after the second failed retry. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Lumen's data retention approval threshold has a 12 percent cap in Q3 after the second failed retry, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "96 percent; when the customer tier is enterprise; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-1a1d4acf26307581", "meaning-doc-decd3317254f6a9c", "meaning-doc-004759327d40ef5f"], "hard_negative_texts": ["Cedar clinical triage memo 62: The escalation limit is capped at 96 percent during Q2, when the customer tier is starter. Archived exceptions are counted as active for audit sampling.", "Cedar clinical triage memo 62: The escalation limit is capped at 93 percent during Q2, when the customer tier is enterprise. The note also says archived exceptions remain active.", "Cedar clinical triage memo 62: The escalation limit target is at least 96 percent during Q2, when the customer tier is enterprise. Teams may exceed it when capacity is low."], "id": "meaning-q-47200b6c3c26f49b", "positive_doc_id": "meaning-doc-8a285580a2fb13f0", "positive_text": "Cedar clinical triage memo 62: The escalation limit is capped at 96 percent during Q2, when the customer tier is enterprise. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Cedar's clinical triage escalation limit has a 96 percent cap in Q2 when the customer tier is enterprise, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "7 days; only when the request is marked external; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-144eef829e198aba", "meaning-doc-a803297b782b9e76", "meaning-doc-f34b6e133768510c"], "hard_negative_texts": ["Garnet data retention memo 63: The approval threshold is capped at 7 days during Q4, for every request regardless of source. Archived exceptions are counted as active for audit sampling.", "Garnet data retention memo 63: The approval threshold is capped at 10 days during Q4, only when the request is marked external. The note also says archived exceptions remain active.", "Garnet data retention memo 63: The approval threshold target is at least 7 days during Q4, only when the request is marked external. Teams may exceed it when capacity is low."], "id": "meaning-q-c32e1a097facd50c", "positive_doc_id": "meaning-doc-7c7e2adfe7854628", "positive_text": "Garnet data retention memo 63: The approval threshold is capped at 7 days during Q4, only when the request is marked external. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Garnet's data retention approval threshold has a 7 days cap in Q4 only when the request is marked external, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "96 hours; when the customer tier is enterprise; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-df228cacac94a72f", "meaning-doc-538b7dcbf0118aac", "meaning-doc-07b2aca2069e6869"], "hard_negative_texts": ["Ember clinical triage memo 64: The rollback window is capped at 96 hours during Q4, after the regional handoff is signed. Archived exceptions are counted as active for audit sampling.", "Ember clinical triage memo 64: The rollback window is capped at 94 hours during Q4, when the customer tier is enterprise. The note also says archived exceptions remain active.", "Ember clinical triage memo 64: The rollback window target is at least 96 hours during Q4, when the customer tier is enterprise. Teams may exceed it when capacity is low."], "id": "meaning-q-cda180348e3abdc6", "positive_doc_id": "meaning-doc-81dd2c1029580dce", "positive_text": "Ember clinical triage memo 64: The rollback window is capped at 96 hours during Q4, when the customer tier is enterprise. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Ember's clinical triage rollback window has a 96 hours cap in Q4 when the customer tier is enterprise, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "7 percent; unless the owner is the platform team; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-d072a146eec02206", "meaning-doc-b4a54ecc3a8e5ee7", "meaning-doc-93b8af5ae030804a"], "hard_negative_texts": ["Ember warehouse operations memo 65: The escalation limit is capped at 7 percent during Q2, after the first successful retry. Archived exceptions are counted as active for audit sampling.", "Ember warehouse operations memo 65: The escalation limit is capped at 4 percent during Q2, unless the owner is the platform team. The note also says archived exceptions remain active.", "Ember warehouse operations memo 65: The escalation limit target is at least 7 percent during Q2, unless the owner is the platform team. Teams may exceed it when capacity is low."], "id": "meaning-q-987cff8c7715d5ce", "positive_doc_id": "meaning-doc-7a8c2ab0a8836a88", "positive_text": "Ember warehouse operations memo 65: The escalation limit is capped at 7 percent during Q2, unless the owner is the platform team. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Ember's warehouse operations escalation limit has a 7 percent cap in Q2 unless the owner is the platform team, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "96 hours; before the regional handoff is signed; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-378cb3cdbd394b60", "meaning-doc-6ee4ea1121f40fc0", "meaning-doc-53e741c3975b5eee"], "hard_negative_texts": ["Garnet warehouse operations memo 66: The sampling rate is capped at 96 hours during Q4, after the regional handoff is signed. Archived exceptions are counted as active for audit sampling.", "Garnet warehouse operations memo 66: The sampling rate is capped at 98 hours during Q4, before the regional handoff is signed. The note also says archived exceptions remain active.", "Garnet warehouse operations memo 66: The sampling rate target is at least 96 hours during Q4, before the regional handoff is signed. Teams may exceed it when capacity is low."], "id": "meaning-q-f41ff99abcb89b89", "positive_doc_id": "meaning-doc-d46ca36833dcc4c7", "positive_text": "Garnet warehouse operations memo 66: The sampling rate is capped at 96 hours during Q4, before the regional handoff is signed. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Garnet's warehouse operations sampling rate has a 96 hours cap in Q4 before the regional handoff is signed, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "72 percent; before the regional handoff is signed; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-3d3b08b1b14e31bb", "meaning-doc-b8463cf3734fad63", "meaning-doc-a19849a57cab9078"], "hard_negative_texts": ["Ember edge caching memo 67: The sampling rate is capped at 72 percent during Q3, after the first successful retry. Archived exceptions are counted as active for audit sampling.", "Ember edge caching memo 67: The sampling rate is capped at 75 percent during Q3, before the regional handoff is signed. The note also says archived exceptions remain active.", "Ember edge caching memo 67: The sampling rate target is at least 72 percent during Q3, before the regional handoff is signed. Teams may exceed it when capacity is low."], "id": "meaning-q-3bd3752a360ba0ab", "positive_doc_id": "meaning-doc-e73a97040e513fa7", "positive_text": "Ember edge caching memo 67: The sampling rate is capped at 72 percent during Q3, before the regional handoff is signed. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Ember's edge caching sampling rate has a 72 percent cap in Q3 before the regional handoff is signed, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "24 requests; after the second failed retry; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-1ed2816d2fa50504", "meaning-doc-4fd3c5416041269a", "meaning-doc-f6034f33c4b5a1bb"], "hard_negative_texts": ["Delta privacy requests memo 68: The review level is capped at 24 requests during Q4, after the first successful retry. Archived exceptions are counted as active for audit sampling.", "Delta privacy requests memo 68: The review level is capped at 26 requests during Q4, after the second failed retry. The note also says archived exceptions remain active.", "Delta privacy requests memo 68: The review level target is at least 24 requests during Q4, after the second failed retry. Teams may exceed it when capacity is low."], "id": "meaning-q-10972b94d674c2bc", "positive_doc_id": "meaning-doc-a29b5799224c0529", "positive_text": "Delta privacy requests memo 68: The review level is capped at 24 requests during Q4, after the second failed retry. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Delta's privacy requests review level has a 24 requests cap in Q4 after the second failed retry, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "7 hours; unless the owner is the platform team; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-8c234cdbd5450567", "meaning-doc-0e429474621b6824", "meaning-doc-4ba485bb31502ec7"], "hard_negative_texts": ["Indigo payments memo 69: The approval threshold is capped at 7 hours during Q3, after the regional handoff is signed. Archived exceptions are counted as active for audit sampling.", "Indigo payments memo 69: The approval threshold is capped at 4 hours during Q3, unless the owner is the platform team. The note also says archived exceptions remain active.", "Indigo payments memo 69: The approval threshold target is at least 7 hours during Q3, unless the owner is the platform team. Teams may exceed it when capacity is low."], "id": "meaning-q-76ba86a4dd895231", "positive_doc_id": "meaning-doc-c9c541db38a8c172", "positive_text": "Indigo payments memo 69: The approval threshold is capped at 7 hours during Q3, unless the owner is the platform team. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Indigo's payments approval threshold has a 7 hours cap in Q3 unless the owner is the platform team, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "48 tickets; before the regional handoff is signed; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-15b2d28cbbe04f59", "meaning-doc-8a326130ffb34c13", "meaning-doc-adddf7a37c489d19"], "hard_negative_texts": ["Basil cloud billing memo 70: The escalation limit is capped at 48 tickets during Q1, for every request regardless of source. Archived exceptions are counted as active for audit sampling.", "Basil cloud billing memo 70: The escalation limit is capped at 46 tickets during Q1, before the regional handoff is signed. The note also says archived exceptions remain active.", "Basil cloud billing memo 70: The escalation limit target is at least 48 tickets during Q1, before the regional handoff is signed. Teams may exceed it when capacity is low."], "id": "meaning-q-f56bcc3204005c96", "positive_doc_id": "meaning-doc-15aeccddb27426b9", "positive_text": "Basil cloud billing memo 70: The escalation limit is capped at 48 tickets during Q1, before the regional handoff is signed. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Basil's cloud billing escalation limit has a 48 tickets cap in Q1 before the regional handoff is signed, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "12 hours; after the second failed retry; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-28441f2016238bc5", "meaning-doc-8bddca297af7fe42", "meaning-doc-3dac80e130c0c37b"], "hard_negative_texts": ["Ember payments memo 71: The payment hold is capped at 12 hours during Q4, after the regional handoff is signed. Archived exceptions are counted as active for audit sampling.", "Ember payments memo 71: The payment hold is capped at 14 hours during Q4, after the second failed retry. The note also says archived exceptions remain active.", "Ember payments memo 71: The payment hold target is at least 12 hours during Q4, after the second failed retry. Teams may exceed it when capacity is low."], "id": "meaning-q-a745463216df5e17", "positive_doc_id": "meaning-doc-bccd04b93eb69e14", "positive_text": "Ember payments memo 71: The payment hold is capped at 12 hours during Q4, after the second failed retry. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Ember's payments payment hold has a 12 hours cap in Q4 after the second failed retry, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "96 minutes; after the second failed retry; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-c568e2507e7b0df2", "meaning-doc-ffaf1e9bc35fbfee", "meaning-doc-e40f98010a9bcfea"], "hard_negative_texts": ["Ember payments memo 72: The rollback window is capped at 96 minutes during Q1, after the regional handoff is signed. Archived exceptions are counted as active for audit sampling.", "Ember payments memo 72: The rollback window is capped at 99 minutes during Q1, after the second failed retry. The note also says archived exceptions remain active.", "Ember payments memo 72: The rollback window target is at least 96 minutes during Q1, after the second failed retry. Teams may exceed it when capacity is low."], "id": "meaning-q-f086e2e78eb0bc60", "positive_doc_id": "meaning-doc-b9d158b173db19a2", "positive_text": "Ember payments memo 72: The rollback window is capped at 96 minutes during Q1, after the second failed retry. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Ember's payments rollback window has a 96 minutes cap in Q1 after the second failed retry, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "48 minutes; before the regional handoff is signed; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-f4197546131be5c7", "meaning-doc-2a20fd0bee9a6ad2", "meaning-doc-b74e9ba2a8b20146"], "hard_negative_texts": ["Lumen release management memo 73: The payment hold is capped at 48 minutes during Q3, when the customer tier is starter. Archived exceptions are counted as active for audit sampling.", "Lumen release management memo 73: The payment hold is capped at 46 minutes during Q3, before the regional handoff is signed. The note also says archived exceptions remain active.", "Lumen release management memo 73: The payment hold target is at least 48 minutes during Q3, before the regional handoff is signed. Teams may exceed it when capacity is low."], "id": "meaning-q-5ebb795e549402d6", "positive_doc_id": "meaning-doc-f34a5eb6186d8b2c", "positive_text": "Lumen release management memo 73: The payment hold is capped at 48 minutes during Q3, before the regional handoff is signed. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Lumen's release management payment hold has a 48 minutes cap in Q3 before the regional handoff is signed, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "12 requests; after the second failed retry; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-35eff34152f69a2a", "meaning-doc-22f7616fc1ee2017", "meaning-doc-cc097b2d78508f30"], "hard_negative_texts": ["Garnet edge caching memo 74: The approval threshold is capped at 12 requests during Q3, only when the owner is the platform team. Archived exceptions are counted as active for audit sampling.", "Garnet edge caching memo 74: The approval threshold is capped at 18 requests during Q3, after the second failed retry. The note also says archived exceptions remain active.", "Garnet edge caching memo 74: The approval threshold target is at least 12 requests during Q3, after the second failed retry. Teams may exceed it when capacity is low."], "id": "meaning-q-5da69b380857d131", "positive_doc_id": "meaning-doc-c24ca05ef2481050", "positive_text": "Garnet edge caching memo 74: The approval threshold is capped at 12 requests during Q3, after the second failed retry. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Garnet's edge caching approval threshold has a 12 requests cap in Q3 after the second failed retry, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "18 requests; unless the owner is the platform team; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-d8c449b274200eb7", "meaning-doc-aad987828dfc3235", "meaning-doc-c09426219b8ec5ec"], "hard_negative_texts": ["Delta privacy requests memo 75: The latency is capped at 18 requests during Q1, after the first successful retry. Archived exceptions are counted as active for audit sampling.", "Delta privacy requests memo 75: The latency is capped at 21 requests during Q1, unless the owner is the platform team. The note also says archived exceptions remain active.", "Delta privacy requests memo 75: The latency target is at least 18 requests during Q1, unless the owner is the platform team. Teams may exceed it when capacity is low."], "id": "meaning-q-c4b3825761b444b6", "positive_doc_id": "meaning-doc-260db72e775d23ac", "positive_text": "Delta privacy requests memo 75: The latency is capped at 18 requests during Q1, unless the owner is the platform team. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Delta's privacy requests latency has a 18 requests cap in Q1 unless the owner is the platform team, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "96 days; when the customer tier is enterprise; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-db5788ed03b81360", "meaning-doc-c81b5a35f3282890", "meaning-doc-639ad23e2b6d9b9a"], "hard_negative_texts": ["Lumen security review memo 76: The rollback window is capped at 96 days during Q2, only when the owner is the platform team. Archived exceptions are counted as active for audit sampling.", "Lumen security review memo 76: The rollback window is capped at 99 days during Q2, when the customer tier is enterprise. The note also says archived exceptions remain active.", "Lumen security review memo 76: The rollback window target is at least 96 days during Q2, when the customer tier is enterprise. Teams may exceed it when capacity is low."], "id": "meaning-q-dc90a786320ba575", "positive_doc_id": "meaning-doc-fdb116e18564641f", "positive_text": "Lumen security review memo 76: The rollback window is capped at 96 days during Q2, when the customer tier is enterprise. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Lumen's security review rollback window has a 96 days cap in Q2 when the customer tier is enterprise, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "36 days; before the regional handoff is signed; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-22fc765b9603e942", "meaning-doc-c255c1d602a3f0f3", "meaning-doc-ad263105a1dfee7e"], "hard_negative_texts": ["Kepler security review memo 77: The retention period is capped at 36 days during Q1, after the regional handoff is signed. Archived exceptions are counted as active for audit sampling.", "Kepler security review memo 77: The retention period is capped at 33 days during Q1, before the regional handoff is signed. The note also says archived exceptions remain active.", "Kepler security review memo 77: The retention period target is at least 36 days during Q1, before the regional handoff is signed. Teams may exceed it when capacity is low."], "id": "meaning-q-ef114b34bb24d190", "positive_doc_id": "meaning-doc-9a4ab7678a96d4e2", "positive_text": "Kepler security review memo 77: The retention period is capped at 36 days during Q1, before the regional handoff is signed. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Kepler's security review retention period has a 36 days cap in Q1 before the regional handoff is signed, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "96 requests; when the customer tier is enterprise; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-988ab6ee87faf3e0", "meaning-doc-de8a4ccbbc78ff20", "meaning-doc-468494b41169edaa"], "hard_negative_texts": ["Cedar edge caching memo 78: The escalation limit is capped at 96 requests during Q3, only when the owner is the platform team. Archived exceptions are counted as active for audit sampling.", "Cedar edge caching memo 78: The escalation limit is capped at 98 requests during Q3, when the customer tier is enterprise. The note also says archived exceptions remain active.", "Cedar edge caching memo 78: The escalation limit target is at least 96 requests during Q3, when the customer tier is enterprise. Teams may exceed it when capacity is low."], "id": "meaning-q-f5774541b3644860", "positive_doc_id": "meaning-doc-085d57e38d59db91", "positive_text": "Cedar edge caching memo 78: The escalation limit is capped at 96 requests during Q3, when the customer tier is enterprise. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Cedar's edge caching escalation limit has a 96 requests cap in Q3 when the customer tier is enterprise, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "96 percent; after the second failed retry; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-31d3471b9a0dd585", "meaning-doc-e830e64ac20a8624", "meaning-doc-a87f319cfef8958e"], "hard_negative_texts": ["Kepler procurement memo 79: The sampling rate is capped at 96 percent during Q2, only when the owner is the platform team. Archived exceptions are counted as active for audit sampling.", "Kepler procurement memo 79: The sampling rate is capped at 98 percent during Q2, after the second failed retry. The note also says archived exceptions remain active.", "Kepler procurement memo 79: The sampling rate target is at least 96 percent during Q2, after the second failed retry. Teams may exceed it when capacity is low."], "id": "meaning-q-4958e38b9c2cd08f", "positive_doc_id": "meaning-doc-1057ddfc826c297c", "positive_text": "Kepler procurement memo 79: The sampling rate is capped at 96 percent during Q2, after the second failed retry. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Kepler's procurement sampling rate has a 96 percent cap in Q2 after the second failed retry, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "24 minutes; before the regional handoff is signed; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-21e687ce57211687", "meaning-doc-07a273eb0267fc32", "meaning-doc-3508499ca1919d17"], "hard_negative_texts": ["Kepler edge caching memo 80: The approval threshold is capped at 24 minutes during Q2, only when the owner is the platform team. Archived exceptions are counted as active for audit sampling.", "Kepler edge caching memo 80: The approval threshold is capped at 27 minutes during Q2, before the regional handoff is signed. The note also says archived exceptions remain active.", "Kepler edge caching memo 80: The approval threshold target is at least 24 minutes during Q2, before the regional handoff is signed. Teams may exceed it when capacity is low."], "id": "meaning-q-a0962e388d7b3a68", "positive_doc_id": "meaning-doc-1945eb83bf85962d", "positive_text": "Kepler edge caching memo 80: The approval threshold is capped at 24 minutes during Q2, before the regional handoff is signed. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Kepler's edge caching approval threshold has a 24 minutes cap in Q2 before the regional handoff is signed, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "36 minutes; unless the owner is the platform team; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-df3372fd82fb75d2", "meaning-doc-eb18edb58d9b1bae", "meaning-doc-9f60ef606ddd2158"], "hard_negative_texts": ["Cedar model deployment memo 81: The rollback window is capped at 36 minutes during Q4, for every request regardless of source. Archived exceptions are counted as active for audit sampling.", "Cedar model deployment memo 81: The rollback window is capped at 42 minutes during Q4, unless the owner is the platform team. The note also says archived exceptions remain active.", "Cedar model deployment memo 81: The rollback window target is at least 36 minutes during Q4, unless the owner is the platform team. Teams may exceed it when capacity is low."], "id": "meaning-q-27978771f06afec5", "positive_doc_id": "meaning-doc-5423f00a00d53ae9", "positive_text": "Cedar model deployment memo 81: The rollback window is capped at 36 minutes during Q4, unless the owner is the platform team. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Cedar's model deployment rollback window has a 36 minutes cap in Q4 unless the owner is the platform team, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "18 percent; only when the request is marked external; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-1bc0da8dc0da3282", "meaning-doc-adc6aa42f87b254c", "meaning-doc-3a543fc24d4bcfd1"], "hard_negative_texts": ["Kepler model deployment memo 82: The error budget is capped at 18 percent during Q1, after the regional handoff is signed. Archived exceptions are counted as active for audit sampling.", "Kepler model deployment memo 82: The error budget is capped at 16 percent during Q1, only when the request is marked external. The note also says archived exceptions remain active.", "Kepler model deployment memo 82: The error budget target is at least 18 percent during Q1, only when the request is marked external. Teams may exceed it when capacity is low."], "id": "meaning-q-1e6c7cee97ebd013", "positive_doc_id": "meaning-doc-0d90e29687e940e6", "positive_text": "Kepler model deployment memo 82: The error budget is capped at 18 percent during Q1, only when the request is marked external. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Kepler's model deployment error budget has a 18 percent cap in Q1 only when the request is marked external, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "36 minutes; unless the owner is the platform team; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-4a1041f71f816a63", "meaning-doc-34641c37287570aa", "meaning-doc-4456424707064b62"], "hard_negative_texts": ["Basil incident response memo 83: The approval threshold is capped at 36 minutes during Q4, after the first successful retry. Archived exceptions are counted as active for audit sampling.", "Basil incident response memo 83: The approval threshold is capped at 34 minutes during Q4, unless the owner is the platform team. The note also says archived exceptions remain active.", "Basil incident response memo 83: The approval threshold target is at least 36 minutes during Q4, unless the owner is the platform team. Teams may exceed it when capacity is low."], "id": "meaning-q-f0bff7147c40e626", "positive_doc_id": "meaning-doc-28dac5dfd6693aec", "positive_text": "Basil incident response memo 83: The approval threshold is capped at 36 minutes during Q4, unless the owner is the platform team. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Basil's incident response approval threshold has a 36 minutes cap in Q4 unless the owner is the platform team, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "7 minutes; when the customer tier is enterprise; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-0eda24ad56524656", "meaning-doc-db7986626840cabc", "meaning-doc-dfe8d1da81678e4c"], "hard_negative_texts": ["Basil cloud billing memo 84: The error budget is capped at 7 minutes during Q1, only when the owner is the platform team. Archived exceptions are counted as active for audit sampling.", "Basil cloud billing memo 84: The error budget is capped at 9 minutes during Q1, when the customer tier is enterprise. The note also says archived exceptions remain active.", "Basil cloud billing memo 84: The error budget target is at least 7 minutes during Q1, when the customer tier is enterprise. Teams may exceed it when capacity is low."], "id": "meaning-q-af00683299f251ed", "positive_doc_id": "meaning-doc-69a21bb8acc7c941", "positive_text": "Basil cloud billing memo 84: The error budget is capped at 7 minutes during Q1, when the customer tier is enterprise. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Basil's cloud billing error budget has a 7 minutes cap in Q1 when the customer tier is enterprise, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "48 tickets; before the regional handoff is signed; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-3c9b581bcd78baea", "meaning-doc-104f5007d8e70429", "meaning-doc-0df756dcbb38f300"], "hard_negative_texts": ["Juniper release management memo 85: The approval threshold is capped at 48 tickets during Q2, only when the owner is the platform team. Archived exceptions are counted as active for audit sampling.", "Juniper release management memo 85: The approval threshold is capped at 45 tickets during Q2, before the regional handoff is signed. The note also says archived exceptions remain active.", "Juniper release management memo 85: The approval threshold target is at least 48 tickets during Q2, before the regional handoff is signed. Teams may exceed it when capacity is low."], "id": "meaning-q-875b4f5fdca4aea5", "positive_doc_id": "meaning-doc-622c6b8a7021f547", "positive_text": "Juniper release management memo 85: The approval threshold is capped at 48 tickets during Q2, before the regional handoff is signed. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Juniper's release management approval threshold has a 48 tickets cap in Q2 before the regional handoff is signed, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "24 tickets; after the second failed retry; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-eec6c11f248cf47c", "meaning-doc-5bb2d131fd8bc026", "meaning-doc-92245d81d469a8a3"], "hard_negative_texts": ["Basil security review memo 86: The review level is capped at 24 tickets during Q1, for every request regardless of source. Archived exceptions are counted as active for audit sampling.", "Basil security review memo 86: The review level is capped at 26 tickets during Q1, after the second failed retry. The note also says archived exceptions remain active.", "Basil security review memo 86: The review level target is at least 24 tickets during Q1, after the second failed retry. Teams may exceed it when capacity is low."], "id": "meaning-q-7a0ee50a790c342c", "positive_doc_id": "meaning-doc-5cc2cba9c577526b", "positive_text": "Basil security review memo 86: The review level is capped at 24 tickets during Q1, after the second failed retry. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Basil's security review review level has a 24 tickets cap in Q1 after the second failed retry, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "72 days; before the regional handoff is signed; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-bf7ee4f66d69490e", "meaning-doc-8ddecff3cf89bc1a", "meaning-doc-549e1dd036b71201"], "hard_negative_texts": ["Juniper warehouse operations memo 87: The payment hold is capped at 72 days during Q2, after the regional handoff is signed. Archived exceptions are counted as active for audit sampling.", "Juniper warehouse operations memo 87: The payment hold is capped at 69 days during Q2, before the regional handoff is signed. The note also says archived exceptions remain active.", "Juniper warehouse operations memo 87: The payment hold target is at least 72 days during Q2, before the regional handoff is signed. Teams may exceed it when capacity is low."], "id": "meaning-q-9e11b7a9b9be4e10", "positive_doc_id": "meaning-doc-f09e6bb2cce02167", "positive_text": "Juniper warehouse operations memo 87: The payment hold is capped at 72 days during Q2, before the regional handoff is signed. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Juniper's warehouse operations payment hold has a 72 days cap in Q2 before the regional handoff is signed, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "18 days; unless the owner is the platform team; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-ea7bc655a532ef80", "meaning-doc-9fe6f49749d8ddc3", "meaning-doc-533e60da0a47aecd"], "hard_negative_texts": ["Ember data retention memo 88: The retention period is capped at 18 days during Q2, only when the owner is the platform team. Archived exceptions are counted as active for audit sampling.", "Ember data retention memo 88: The retention period is capped at 16 days during Q2, unless the owner is the platform team. The note also says archived exceptions remain active.", "Ember data retention memo 88: The retention period target is at least 18 days during Q2, unless the owner is the platform team. Teams may exceed it when capacity is low."], "id": "meaning-q-b48079bde7b1e337", "positive_doc_id": "meaning-doc-0ce13f7b59de99a6", "positive_text": "Ember data retention memo 88: The retention period is capped at 18 days during Q2, unless the owner is the platform team. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Ember's data retention retention period has a 18 days cap in Q2 unless the owner is the platform team, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "7 requests; unless the owner is the platform team; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-ca503957a2cc42a4", "meaning-doc-27fda0aa587cbd3e", "meaning-doc-10a91aa3449e1c76"], "hard_negative_texts": ["Aurora security review memo 89: The cache lifetime is capped at 7 requests during Q1, for every request regardless of source. Archived exceptions are counted as active for audit sampling.", "Aurora security review memo 89: The cache lifetime is capped at 4 requests during Q1, unless the owner is the platform team. The note also says archived exceptions remain active.", "Aurora security review memo 89: The cache lifetime target is at least 7 requests during Q1, unless the owner is the platform team. Teams may exceed it when capacity is low."], "id": "meaning-q-3be6dee3f39fee10", "positive_doc_id": "meaning-doc-ab61c9911879e3e5", "positive_text": "Aurora security review memo 89: The cache lifetime is capped at 7 requests during Q1, unless the owner is the platform team. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Aurora's security review cache lifetime has a 7 requests cap in Q1 unless the owner is the platform team, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "36 hours; unless the owner is the platform team; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-8817e1378d610af5", "meaning-doc-4770b87aba90e0b9", "meaning-doc-0a46a0ce1347662e"], "hard_negative_texts": ["Fjord model deployment memo 90: The sampling rate is capped at 36 hours during Q1, after the regional handoff is signed. Archived exceptions are counted as active for audit sampling.", "Fjord model deployment memo 90: The sampling rate is capped at 34 hours during Q1, unless the owner is the platform team. The note also says archived exceptions remain active.", "Fjord model deployment memo 90: The sampling rate target is at least 36 hours during Q1, unless the owner is the platform team. Teams may exceed it when capacity is low."], "id": "meaning-q-fd0192fac012074c", "positive_doc_id": "meaning-doc-eef6d1152cc4c91d", "positive_text": "Fjord model deployment memo 90: The sampling rate is capped at 36 hours during Q1, unless the owner is the platform team. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Fjord's model deployment sampling rate has a 36 hours cap in Q1 unless the owner is the platform team, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "12 tickets; before the regional handoff is signed; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-fd9f34fb7cf46741", "meaning-doc-409813ef3c75ed93", "meaning-doc-9e283fac689b58a5"], "hard_negative_texts": ["Harbor warehouse operations memo 91: The approval threshold is capped at 12 tickets during Q2, only when the owner is the platform team. Archived exceptions are counted as active for audit sampling.", "Harbor warehouse operations memo 91: The approval threshold is capped at 14 tickets during Q2, before the regional handoff is signed. The note also says archived exceptions remain active.", "Harbor warehouse operations memo 91: The approval threshold target is at least 12 tickets during Q2, before the regional handoff is signed. Teams may exceed it when capacity is low."], "id": "meaning-q-fb653419ffd5b450", "positive_doc_id": "meaning-doc-b7e5292109226982", "positive_text": "Harbor warehouse operations memo 91: The approval threshold is capped at 12 tickets during Q2, before the regional handoff is signed. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Harbor's warehouse operations approval threshold has a 12 tickets cap in Q2 before the regional handoff is signed, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "72 days; unless the owner is the platform team; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-060fd14553671afa", "meaning-doc-06ca9d037558d374", "meaning-doc-3b87a222e10957bd"], "hard_negative_texts": ["Juniper security review memo 92: The review level is capped at 72 days during Q4, only when the owner is the platform team. Archived exceptions are counted as active for audit sampling.", "Juniper security review memo 92: The review level is capped at 75 days during Q4, unless the owner is the platform team. The note also says archived exceptions remain active.", "Juniper security review memo 92: The review level target is at least 72 days during Q4, unless the owner is the platform team. Teams may exceed it when capacity is low."], "id": "meaning-q-0dcf1c0c837ed019", "positive_doc_id": "meaning-doc-a475c27a06f0dd72", "positive_text": "Juniper security review memo 92: The review level is capped at 72 days during Q4, unless the owner is the platform team. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Juniper's security review review level has a 72 days cap in Q4 unless the owner is the platform team, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "72 tickets; before the regional handoff is signed; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-0638156b16dca192", "meaning-doc-aadc51e7d5479b4f", "meaning-doc-6abb07e06fe71de7"], "hard_negative_texts": ["Ember privacy requests memo 93: The sampling rate is capped at 72 tickets during Q3, after the regional handoff is signed. Archived exceptions are counted as active for audit sampling.", "Ember privacy requests memo 93: The sampling rate is capped at 75 tickets during Q3, before the regional handoff is signed. The note also says archived exceptions remain active.", "Ember privacy requests memo 93: The sampling rate target is at least 72 tickets during Q3, before the regional handoff is signed. Teams may exceed it when capacity is low."], "id": "meaning-q-dd1a0dc515b4cdb5", "positive_doc_id": "meaning-doc-f02ccb85427f1af9", "positive_text": "Ember privacy requests memo 93: The sampling rate is capped at 72 tickets during Q3, before the regional handoff is signed. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Ember's privacy requests sampling rate has a 72 tickets cap in Q3 before the regional handoff is signed, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "7 percent; after the second failed retry; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-6f95617d887e31af", "meaning-doc-7f5485bb7e9add28", "meaning-doc-51846f043bebbfc4"], "hard_negative_texts": ["Aurora release management memo 94: The rollback window is capped at 7 percent during Q3, after the regional handoff is signed. Archived exceptions are counted as active for audit sampling.", "Aurora release management memo 94: The rollback window is capped at 13 percent during Q3, after the second failed retry. The note also says archived exceptions remain active.", "Aurora release management memo 94: The rollback window target is at least 7 percent during Q3, after the second failed retry. Teams may exceed it when capacity is low."], "id": "meaning-q-5cb36fc589b939fb", "positive_doc_id": "meaning-doc-15ca8d7bdbd5ca26", "positive_text": "Aurora release management memo 94: The rollback window is capped at 7 percent during Q3, after the second failed retry. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Aurora's release management rollback window has a 7 percent cap in Q3 after the second failed retry, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "36 tickets; before the regional handoff is signed; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-2baf136f82faeaf5", "meaning-doc-a77f1a989d271b28", "meaning-doc-8a73632fdb236c7f"], "hard_negative_texts": ["Lumen payments memo 95: The cache lifetime is capped at 36 tickets during Q3, when the customer tier is starter. Archived exceptions are counted as active for audit sampling.", "Lumen payments memo 95: The cache lifetime is capped at 38 tickets during Q3, before the regional handoff is signed. The note also says archived exceptions remain active.", "Lumen payments memo 95: The cache lifetime target is at least 36 tickets during Q3, before the regional handoff is signed. Teams may exceed it when capacity is low."], "id": "meaning-q-d32d9a814db22d24", "positive_doc_id": "meaning-doc-28f3d3be0620f387", "positive_text": "Lumen payments memo 95: The cache lifetime is capped at 36 tickets during Q3, before the regional handoff is signed. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Lumen's payments cache lifetime has a 36 tickets cap in Q3 before the regional handoff is signed, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "12 tickets; only when the request is marked external; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-422cb11117849556", "meaning-doc-403f513c2ca779c5", "meaning-doc-bbec79065db26f0e"], "hard_negative_texts": ["Delta cloud billing memo 96: The payment hold is capped at 12 tickets during Q4, after the regional handoff is signed. Archived exceptions are counted as active for audit sampling.", "Delta cloud billing memo 96: The payment hold is capped at 14 tickets during Q4, only when the request is marked external. The note also says archived exceptions remain active.", "Delta cloud billing memo 96: The payment hold target is at least 12 tickets during Q4, only when the request is marked external. Teams may exceed it when capacity is low."], "id": "meaning-q-8236140131626908", "positive_doc_id": "meaning-doc-667505a83c3083f4", "positive_text": "Delta cloud billing memo 96: The payment hold is capped at 12 tickets during Q4, only when the request is marked external. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Delta's cloud billing payment hold has a 12 tickets cap in Q4 only when the request is marked external, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "7 requests; after the second failed retry; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-72ddf14272df93e0", "meaning-doc-5e749b970cac5806", "meaning-doc-63b23a889c9d53dc"], "hard_negative_texts": ["Ember clinical triage memo 97: The sampling rate is capped at 7 requests during Q3, only when the owner is the platform team. Archived exceptions are counted as active for audit sampling.", "Ember clinical triage memo 97: The sampling rate is capped at 10 requests during Q3, after the second failed retry. The note also says archived exceptions remain active.", "Ember clinical triage memo 97: The sampling rate target is at least 7 requests during Q3, after the second failed retry. Teams may exceed it when capacity is low."], "id": "meaning-q-f13cb7a29d08c59d", "positive_doc_id": "meaning-doc-bfacfa9046f08b37", "positive_text": "Ember clinical triage memo 97: The sampling rate is capped at 7 requests during Q3, after the second failed retry. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Ember's clinical triage sampling rate has a 7 requests cap in Q3 after the second failed retry, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "48 percent; only when the request is marked external; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-6927ad2ecac60ace", "meaning-doc-afc52285dfdae29c", "meaning-doc-470c180dc841a372"], "hard_negative_texts": ["Garnet incident response memo 98: The error budget is capped at 48 percent during Q1, only when the owner is the platform team. Archived exceptions are counted as active for audit sampling.", "Garnet incident response memo 98: The error budget is capped at 50 percent during Q1, only when the request is marked external. The note also says archived exceptions remain active.", "Garnet incident response memo 98: The error budget target is at least 48 percent during Q1, only when the request is marked external. Teams may exceed it when capacity is low."], "id": "meaning-q-0914f92a77b5645a", "positive_doc_id": "meaning-doc-2a6bd70e5cad5cc1", "positive_text": "Garnet incident response memo 98: The error budget is capped at 48 percent during Q1, only when the request is marked external. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Garnet's incident response error budget has a 48 percent cap in Q1 only when the request is marked external, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "12 tickets; when the customer tier is enterprise; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-59a3402ef136cc72", "meaning-doc-2d1f58c068dc39ce", "meaning-doc-e870ec0475755e5f"], "hard_negative_texts": ["Lumen procurement memo 99: The cache lifetime is capped at 12 tickets during Q2, after the first successful retry. Archived exceptions are counted as active for audit sampling.", "Lumen procurement memo 99: The cache lifetime is capped at 10 tickets during Q2, when the customer tier is enterprise. The note also says archived exceptions remain active.", "Lumen procurement memo 99: The cache lifetime target is at least 12 tickets during Q2, when the customer tier is enterprise. Teams may exceed it when capacity is low."], "id": "meaning-q-a4af93f74f7f3ae7", "positive_doc_id": "meaning-doc-84b5e79e64f6bc15", "positive_text": "Lumen procurement memo 99: The cache lifetime is capped at 12 tickets during Q2, when the customer tier is enterprise. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Lumen's procurement cache lifetime has a 12 tickets cap in Q2 when the customer tier is enterprise, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "72 days; after the second failed retry; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-f661fb637c7ec0dc", "meaning-doc-df12da6212ea0cd2", "meaning-doc-6da9cdca57acb841"], "hard_negative_texts": ["Indigo payments memo 100: The payment hold is capped at 72 days during Q3, only when the owner is the platform team. Archived exceptions are counted as active for audit sampling.", "Indigo payments memo 100: The payment hold is capped at 75 days during Q3, after the second failed retry. The note also says archived exceptions remain active.", "Indigo payments memo 100: The payment hold target is at least 72 days during Q3, after the second failed retry. Teams may exceed it when capacity is low."], "id": "meaning-q-079d53de28378fb2", "positive_doc_id": "meaning-doc-92956242d62a62d7", "positive_text": "Indigo payments memo 100: The payment hold is capped at 72 days during Q3, after the second failed retry. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Indigo's payments payment hold has a 72 days cap in Q3 after the second failed retry, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "18 minutes; before the regional handoff is signed; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-9e593388075d6a66", "meaning-doc-ffa5ba36183672fd", "meaning-doc-c6c45ba4e664fd4c"], "hard_negative_texts": ["Lumen privacy requests memo 101: The escalation limit is capped at 18 minutes during Q1, after the first successful retry. Archived exceptions are counted as active for audit sampling.", "Lumen privacy requests memo 101: The escalation limit is capped at 21 minutes during Q1, before the regional handoff is signed. The note also says archived exceptions remain active.", "Lumen privacy requests memo 101: The escalation limit target is at least 18 minutes during Q1, before the regional handoff is signed. Teams may exceed it when capacity is low."], "id": "meaning-q-4808a94f301059d4", "positive_doc_id": "meaning-doc-9467b6ef9236ae80", "positive_text": "Lumen privacy requests memo 101: The escalation limit is capped at 18 minutes during Q1, before the regional handoff is signed. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Lumen's privacy requests escalation limit has a 18 minutes cap in Q1 before the regional handoff is signed, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "12 tickets; only when the request is marked external; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-b840f05f156376e1", "meaning-doc-f90342cff688c076", "meaning-doc-b2e8845c3a409776"], "hard_negative_texts": ["Basil security review memo 102: The rollback window is capped at 12 tickets during Q2, after the first successful retry. Archived exceptions are counted as active for audit sampling.", "Basil security review memo 102: The rollback window is capped at 10 tickets during Q2, only when the request is marked external. The note also says archived exceptions remain active.", "Basil security review memo 102: The rollback window target is at least 12 tickets during Q2, only when the request is marked external. Teams may exceed it when capacity is low."], "id": "meaning-q-d5d1c28ab66cc824", "positive_doc_id": "meaning-doc-1a227ef61420402e", "positive_text": "Basil security review memo 102: The rollback window is capped at 12 tickets during Q2, only when the request is marked external. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Basil's security review rollback window has a 12 tickets cap in Q2 only when the request is marked external, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "12 days; when the customer tier is enterprise; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-c7f282c0174581b4", "meaning-doc-fd86c0224b979bb6", "meaning-doc-73cd99cb8d84248e"], "hard_negative_texts": ["Basil cloud billing memo 103: The review level is capped at 12 days during Q2, after the first successful retry. Archived exceptions are counted as active for audit sampling.", "Basil cloud billing memo 103: The review level is capped at 15 days during Q2, when the customer tier is enterprise. The note also says archived exceptions remain active.", "Basil cloud billing memo 103: The review level target is at least 12 days during Q2, when the customer tier is enterprise. Teams may exceed it when capacity is low."], "id": "meaning-q-7e59894b8f39b46d", "positive_doc_id": "meaning-doc-82bd0620eeaffeee", "positive_text": "Basil cloud billing memo 103: The review level is capped at 12 days during Q2, when the customer tier is enterprise. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Basil's cloud billing review level has a 12 days cap in Q2 when the customer tier is enterprise, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "12 minutes; after the second failed retry; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-cd099fb062b34253", "meaning-doc-ff477558d5403f0f", "meaning-doc-80de70591f5f96be"], "hard_negative_texts": ["Juniper data retention memo 104: The approval threshold is capped at 12 minutes during Q4, only when the owner is the platform team. Archived exceptions are counted as active for audit sampling.", "Juniper data retention memo 104: The approval threshold is capped at 9 minutes during Q4, after the second failed retry. The note also says archived exceptions remain active.", "Juniper data retention memo 104: The approval threshold target is at least 12 minutes during Q4, after the second failed retry. Teams may exceed it when capacity is low."], "id": "meaning-q-c5187097b6046b13", "positive_doc_id": "meaning-doc-cf05ee365d256965", "positive_text": "Juniper data retention memo 104: The approval threshold is capped at 12 minutes during Q4, after the second failed retry. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Juniper's data retention approval threshold has a 12 minutes cap in Q4 after the second failed retry, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "48 days; only when the request is marked external; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-091aa29848108d2f", "meaning-doc-8089fdc64feae26b", "meaning-doc-1089813c9fae3b5a"], "hard_negative_texts": ["Cedar procurement memo 105: The error budget is capped at 48 days during Q4, only when the owner is the platform team. Archived exceptions are counted as active for audit sampling.", "Cedar procurement memo 105: The error budget is capped at 51 days during Q4, only when the request is marked external. The note also says archived exceptions remain active.", "Cedar procurement memo 105: The error budget target is at least 48 days during Q4, only when the request is marked external. Teams may exceed it when capacity is low."], "id": "meaning-q-da7f2996b4ca0342", "positive_doc_id": "meaning-doc-3d55633b772f18cd", "positive_text": "Cedar procurement memo 105: The error budget is capped at 48 days during Q4, only when the request is marked external. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Cedar's procurement error budget has a 48 days cap in Q4 only when the request is marked external, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "96 requests; before the regional handoff is signed; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-f089d4907b4f83cb", "meaning-doc-d1e8da42974d2b70", "meaning-doc-94483544f654ace0"], "hard_negative_texts": ["Cedar incident response memo 106: The rollback window is capped at 96 requests during Q4, only when the owner is the platform team. Archived exceptions are counted as active for audit sampling.", "Cedar incident response memo 106: The rollback window is capped at 99 requests during Q4, before the regional handoff is signed. The note also says archived exceptions remain active.", "Cedar incident response memo 106: The rollback window target is at least 96 requests during Q4, before the regional handoff is signed. Teams may exceed it when capacity is low."], "id": "meaning-q-775934c02b848647", "positive_doc_id": "meaning-doc-2e6c1726664a715b", "positive_text": "Cedar incident response memo 106: The rollback window is capped at 96 requests during Q4, before the regional handoff is signed. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Cedar's incident response rollback window has a 96 requests cap in Q4 before the regional handoff is signed, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "12 hours; after the second failed retry; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-615f6280385089fe", "meaning-doc-612090cf11f9504f", "meaning-doc-cadafa059acd7437"], "hard_negative_texts": ["Indigo warehouse operations memo 107: The error budget is capped at 12 hours during Q1, for every request regardless of source. Archived exceptions are counted as active for audit sampling.", "Indigo warehouse operations memo 107: The error budget is capped at 9 hours during Q1, after the second failed retry. The note also says archived exceptions remain active.", "Indigo warehouse operations memo 107: The error budget target is at least 12 hours during Q1, after the second failed retry. Teams may exceed it when capacity is low."], "id": "meaning-q-c0975315adfc4ee8", "positive_doc_id": "meaning-doc-c10a1fd22b311101", "positive_text": "Indigo warehouse operations memo 107: The error budget is capped at 12 hours during Q1, after the second failed retry. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Indigo's warehouse operations error budget has a 12 hours cap in Q1 after the second failed retry, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "72 days; before the regional handoff is signed; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-584eacac44104565", "meaning-doc-7629e4e43ddf1c21", "meaning-doc-9310523384a26945"], "hard_negative_texts": ["Ember privacy requests memo 108: The approval threshold is capped at 72 days during Q2, for every request regardless of source. Archived exceptions are counted as active for audit sampling.", "Ember privacy requests memo 108: The approval threshold is capped at 75 days during Q2, before the regional handoff is signed. The note also says archived exceptions remain active.", "Ember privacy requests memo 108: The approval threshold target is at least 72 days during Q2, before the regional handoff is signed. Teams may exceed it when capacity is low."], "id": "meaning-q-daf9712ace64e007", "positive_doc_id": "meaning-doc-dbdc69942abd95e4", "positive_text": "Ember privacy requests memo 108: The approval threshold is capped at 72 days during Q2, before the regional handoff is signed. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Ember's privacy requests approval threshold has a 72 days cap in Q2 before the regional handoff is signed, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "24 requests; only when the request is marked external; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-29f77577e6de8527", "meaning-doc-d8560ca5e0053dc1", "meaning-doc-79e76838247a47fb"], "hard_negative_texts": ["Cedar security review memo 109: The cache lifetime is capped at 24 requests during Q2, when the customer tier is starter. Archived exceptions are counted as active for audit sampling.", "Cedar security review memo 109: The cache lifetime is capped at 27 requests during Q2, only when the request is marked external. The note also says archived exceptions remain active.", "Cedar security review memo 109: The cache lifetime target is at least 24 requests during Q2, only when the request is marked external. Teams may exceed it when capacity is low."], "id": "meaning-q-3d8820ae0c38f94e", "positive_doc_id": "meaning-doc-c5a72100ae7b893f", "positive_text": "Cedar security review memo 109: The cache lifetime is capped at 24 requests during Q2, only when the request is marked external. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Cedar's security review cache lifetime has a 24 requests cap in Q2 only when the request is marked external, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "7 days; after the second failed retry; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-abb0babb4d13d14b", "meaning-doc-9562ebff96327899", "meaning-doc-377c22d9c9a25623"], "hard_negative_texts": ["Aurora edge caching memo 110: The retention period is capped at 7 days during Q3, for every request regardless of source. Archived exceptions are counted as active for audit sampling.", "Aurora edge caching memo 110: The retention period is capped at 13 days during Q3, after the second failed retry. The note also says archived exceptions remain active.", "Aurora edge caching memo 110: The retention period target is at least 7 days during Q3, after the second failed retry. Teams may exceed it when capacity is low."], "id": "meaning-q-6ffe665efa02cc31", "positive_doc_id": "meaning-doc-477bc4d58ec23ed9", "positive_text": "Aurora edge caching memo 110: The retention period is capped at 7 days during Q3, after the second failed retry. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Aurora's edge caching retention period has a 7 days cap in Q3 after the second failed retry, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "24 requests; only when the request is marked external; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-38d29745012ac07e", "meaning-doc-b67cce0197f23433", "meaning-doc-67ca4db8bf590414"], "hard_negative_texts": ["Indigo payments memo 111: The cache lifetime is capped at 24 requests during Q2, after the first successful retry. Archived exceptions are counted as active for audit sampling.", "Indigo payments memo 111: The cache lifetime is capped at 22 requests during Q2, only when the request is marked external. The note also says archived exceptions remain active.", "Indigo payments memo 111: The cache lifetime target is at least 24 requests during Q2, only when the request is marked external. Teams may exceed it when capacity is low."], "id": "meaning-q-4569f7c688f79e46", "positive_doc_id": "meaning-doc-efbdea1b80bab98d", "positive_text": "Indigo payments memo 111: The cache lifetime is capped at 24 requests during Q2, only when the request is marked external. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Indigo's payments cache lifetime has a 24 requests cap in Q2 only when the request is marked external, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "18 days; when the customer tier is enterprise; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-7975345d115fe162", "meaning-doc-c45ec9f2ba19ac18", "meaning-doc-e51ef1a30eeb02b2"], "hard_negative_texts": ["Fjord payments memo 112: The cache lifetime is capped at 18 days during Q4, when the customer tier is starter. Archived exceptions are counted as active for audit sampling.", "Fjord payments memo 112: The cache lifetime is capped at 16 days during Q4, when the customer tier is enterprise. The note also says archived exceptions remain active.", "Fjord payments memo 112: The cache lifetime target is at least 18 days during Q4, when the customer tier is enterprise. Teams may exceed it when capacity is low."], "id": "meaning-q-7e1c3bbd3bb74913", "positive_doc_id": "meaning-doc-42d9a0a37a5d206d", "positive_text": "Fjord payments memo 112: The cache lifetime is capped at 18 days during Q4, when the customer tier is enterprise. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Fjord's payments cache lifetime has a 18 days cap in Q4 when the customer tier is enterprise, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "96 minutes; before the regional handoff is signed; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-3d1aa3c322bae095", "meaning-doc-392e3c10f5a84ad7", "meaning-doc-e44feb332ec72dd7"], "hard_negative_texts": ["Delta privacy requests memo 113: The payment hold is capped at 96 minutes during Q3, when the customer tier is starter. Archived exceptions are counted as active for audit sampling.", "Delta privacy requests memo 113: The payment hold is capped at 93 minutes during Q3, before the regional handoff is signed. The note also says archived exceptions remain active.", "Delta privacy requests memo 113: The payment hold target is at least 96 minutes during Q3, before the regional handoff is signed. Teams may exceed it when capacity is low."], "id": "meaning-q-b695e7e078054558", "positive_doc_id": "meaning-doc-af72a6f4c69c6afc", "positive_text": "Delta privacy requests memo 113: The payment hold is capped at 96 minutes during Q3, before the regional handoff is signed. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Delta's privacy requests payment hold has a 96 minutes cap in Q3 before the regional handoff is signed, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "36 percent; only when the request is marked external; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-6d2cbae5e5f06c05", "meaning-doc-9f3eddefe56154df", "meaning-doc-908d91f76fda646f"], "hard_negative_texts": ["Cedar edge caching memo 114: The review level is capped at 36 percent during Q1, only when the owner is the platform team. Archived exceptions are counted as active for audit sampling.", "Cedar edge caching memo 114: The review level is capped at 34 percent during Q1, only when the request is marked external. The note also says archived exceptions remain active.", "Cedar edge caching memo 114: The review level target is at least 36 percent during Q1, only when the request is marked external. Teams may exceed it when capacity is low."], "id": "meaning-q-e63d7397d30ab36a", "positive_doc_id": "meaning-doc-b721a7214585c633", "positive_text": "Cedar edge caching memo 114: The review level is capped at 36 percent during Q1, only when the request is marked external. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Cedar's edge caching review level has a 36 percent cap in Q1 only when the request is marked external, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "36 minutes; when the customer tier is enterprise; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-a9aad3491a729b18", "meaning-doc-fd5f4bfe34df1469", "meaning-doc-68b38c38c46f6013"], "hard_negative_texts": ["Cedar cloud billing memo 115: The retention period is capped at 36 minutes during Q3, when the customer tier is starter. Archived exceptions are counted as active for audit sampling.", "Cedar cloud billing memo 115: The retention period is capped at 33 minutes during Q3, when the customer tier is enterprise. The note also says archived exceptions remain active.", "Cedar cloud billing memo 115: The retention period target is at least 36 minutes during Q3, when the customer tier is enterprise. Teams may exceed it when capacity is low."], "id": "meaning-q-3129c174df5c94b5", "positive_doc_id": "meaning-doc-a0df873427984500", "positive_text": "Cedar cloud billing memo 115: The retention period is capped at 36 minutes during Q3, when the customer tier is enterprise. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Cedar's cloud billing retention period has a 36 minutes cap in Q3 when the customer tier is enterprise, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "12 requests; when the customer tier is enterprise; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-cb3aaa0a868f7e8c", "meaning-doc-20220718c00e7333", "meaning-doc-502145e9c7435697"], "hard_negative_texts": ["Harbor warehouse operations memo 116: The rollback window is capped at 12 requests during Q2, for every request regardless of source. Archived exceptions are counted as active for audit sampling.", "Harbor warehouse operations memo 116: The rollback window is capped at 15 requests during Q2, when the customer tier is enterprise. The note also says archived exceptions remain active.", "Harbor warehouse operations memo 116: The rollback window target is at least 12 requests during Q2, when the customer tier is enterprise. Teams may exceed it when capacity is low."], "id": "meaning-q-cf7a6e713143f7f7", "positive_doc_id": "meaning-doc-48f50a430d99ba2d", "positive_text": "Harbor warehouse operations memo 116: The rollback window is capped at 12 requests during Q2, when the customer tier is enterprise. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Harbor's warehouse operations rollback window has a 12 requests cap in Q2 when the customer tier is enterprise, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "7 percent; before the regional handoff is signed; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-313e0d737cc23acd", "meaning-doc-4e21f1452044ac7b", "meaning-doc-469f4aafba12560a"], "hard_negative_texts": ["Lumen incident response memo 117: The payment hold is capped at 7 percent during Q3, after the regional handoff is signed. Archived exceptions are counted as active for audit sampling.", "Lumen incident response memo 117: The payment hold is capped at 5 percent during Q3, before the regional handoff is signed. The note also says archived exceptions remain active.", "Lumen incident response memo 117: The payment hold target is at least 7 percent during Q3, before the regional handoff is signed. Teams may exceed it when capacity is low."], "id": "meaning-q-ea82f9889e2ba59f", "positive_doc_id": "meaning-doc-19bf613275a70c69", "positive_text": "Lumen incident response memo 117: The payment hold is capped at 7 percent during Q3, before the regional handoff is signed. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Lumen's incident response payment hold has a 7 percent cap in Q3 before the regional handoff is signed, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "12 percent; unless the owner is the platform team; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-030ee3439fa3ccd1", "meaning-doc-3a49a19b56af8b81", "meaning-doc-ae5fe795e384fed6"], "hard_negative_texts": ["Fjord procurement memo 118: The error budget is capped at 12 percent during Q1, after the regional handoff is signed. Archived exceptions are counted as active for audit sampling.", "Fjord procurement memo 118: The error budget is capped at 15 percent during Q1, unless the owner is the platform team. The note also says archived exceptions remain active.", "Fjord procurement memo 118: The error budget target is at least 12 percent during Q1, unless the owner is the platform team. Teams may exceed it when capacity is low."], "id": "meaning-q-60028cd5b862f488", "positive_doc_id": "meaning-doc-a19fc21366671204", "positive_text": "Fjord procurement memo 118: The error budget is capped at 12 percent during Q1, unless the owner is the platform team. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Fjord's procurement error budget has a 12 percent cap in Q1 unless the owner is the platform team, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "36 percent; after the second failed retry; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-d5a000b25af2528a", "meaning-doc-749813938878c4cb", "meaning-doc-83eeaa57b10f7c15"], "hard_negative_texts": ["Indigo data retention memo 119: The retention period is capped at 36 percent during Q3, after the first successful retry. Archived exceptions are counted as active for audit sampling.", "Indigo data retention memo 119: The retention period is capped at 34 percent during Q3, after the second failed retry. The note also says archived exceptions remain active.", "Indigo data retention memo 119: The retention period target is at least 36 percent during Q3, after the second failed retry. Teams may exceed it when capacity is low."], "id": "meaning-q-0daa33d6264ad0b8", "positive_doc_id": "meaning-doc-ba35908d951ec12b", "positive_text": "Indigo data retention memo 119: The retention period is capped at 36 percent during Q3, after the second failed retry. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Indigo's data retention retention period has a 36 percent cap in Q3 after the second failed retry, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "96 requests; only when the request is marked external; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-bfc4eeffd7b86863", "meaning-doc-cd02be182334da2e", "meaning-doc-9ccc78c734cf08b3"], "hard_negative_texts": ["Harbor procurement memo 120: The escalation limit is capped at 96 requests during Q3, only when the owner is the platform team. Archived exceptions are counted as active for audit sampling.", "Harbor procurement memo 120: The escalation limit is capped at 98 requests during Q3, only when the request is marked external. The note also says archived exceptions remain active.", "Harbor procurement memo 120: The escalation limit target is at least 96 requests during Q3, only when the request is marked external. Teams may exceed it when capacity is low."], "id": "meaning-q-a467a5c5408265c2", "positive_doc_id": "meaning-doc-17f5908a256f960f", "positive_text": "Harbor procurement memo 120: The escalation limit is capped at 96 requests during Q3, only when the request is marked external. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Harbor's procurement escalation limit has a 96 requests cap in Q3 only when the request is marked external, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "96 minutes; before the regional handoff is signed; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-24f11bc8585d5965", "meaning-doc-ce18f8be2cecd757", "meaning-doc-648e15950353674c"], "hard_negative_texts": ["Kepler procurement memo 121: The payment hold is capped at 96 minutes during Q2, for every request regardless of source. Archived exceptions are counted as active for audit sampling.", "Kepler procurement memo 121: The payment hold is capped at 99 minutes during Q2, before the regional handoff is signed. The note also says archived exceptions remain active.", "Kepler procurement memo 121: The payment hold target is at least 96 minutes during Q2, before the regional handoff is signed. Teams may exceed it when capacity is low."], "id": "meaning-q-a1f59689d51b13a5", "positive_doc_id": "meaning-doc-48f389066f2cc6f5", "positive_text": "Kepler procurement memo 121: The payment hold is capped at 96 minutes during Q2, before the regional handoff is signed. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Kepler's procurement payment hold has a 96 minutes cap in Q2 before the regional handoff is signed, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "24 percent; only when the request is marked external; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-7320a9058835dc2b", "meaning-doc-766f55e6105cf503", "meaning-doc-76f75db7b5ca5009"], "hard_negative_texts": ["Garnet security review memo 122: The sampling rate is capped at 24 percent during Q2, for every request regardless of source. Archived exceptions are counted as active for audit sampling.", "Garnet security review memo 122: The sampling rate is capped at 30 percent during Q2, only when the request is marked external. The note also says archived exceptions remain active.", "Garnet security review memo 122: The sampling rate target is at least 24 percent during Q2, only when the request is marked external. Teams may exceed it when capacity is low."], "id": "meaning-q-f826dc26742a3800", "positive_doc_id": "meaning-doc-85e7bbbb2f9c2795", "positive_text": "Garnet security review memo 122: The sampling rate is capped at 24 percent during Q2, only when the request is marked external. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Garnet's security review sampling rate has a 24 percent cap in Q2 only when the request is marked external, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "72 tickets; when the customer tier is enterprise; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-e71142d6eda632b7", "meaning-doc-b43fab0ac159f6e9", "meaning-doc-46e23b570f023356"], "hard_negative_texts": ["Ember procurement memo 123: The retention period is capped at 72 tickets during Q2, when the customer tier is starter. Archived exceptions are counted as active for audit sampling.", "Ember procurement memo 123: The retention period is capped at 78 tickets during Q2, when the customer tier is enterprise. The note also says archived exceptions remain active.", "Ember procurement memo 123: The retention period target is at least 72 tickets during Q2, when the customer tier is enterprise. Teams may exceed it when capacity is low."], "id": "meaning-q-9866455677a41ab0", "positive_doc_id": "meaning-doc-c2bfdda913962c8e", "positive_text": "Ember procurement memo 123: The retention period is capped at 72 tickets during Q2, when the customer tier is enterprise. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Ember's procurement retention period has a 72 tickets cap in Q2 when the customer tier is enterprise, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "36 hours; when the customer tier is enterprise; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-e3865a7822a655a1", "meaning-doc-cfe2ad96d6c8660d", "meaning-doc-df57a524cc04b35f"], "hard_negative_texts": ["Kepler warehouse operations memo 124: The review level is capped at 36 hours during Q3, after the regional handoff is signed. Archived exceptions are counted as active for audit sampling.", "Kepler warehouse operations memo 124: The review level is capped at 33 hours during Q3, when the customer tier is enterprise. The note also says archived exceptions remain active.", "Kepler warehouse operations memo 124: The review level target is at least 36 hours during Q3, when the customer tier is enterprise. Teams may exceed it when capacity is low."], "id": "meaning-q-b54adc60f6f58df8", "positive_doc_id": "meaning-doc-38f77e7e033dd3e9", "positive_text": "Kepler warehouse operations memo 124: The review level is capped at 36 hours during Q3, when the customer tier is enterprise. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Kepler's warehouse operations review level has a 36 hours cap in Q3 when the customer tier is enterprise, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "12 requests; when the customer tier is enterprise; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-e0ab40828b2040a5", "meaning-doc-c975200d04784e01", "meaning-doc-601658d0e6d5ea69"], "hard_negative_texts": ["Cedar release management memo 125: The payment hold is capped at 12 requests during Q3, for every request regardless of source. Archived exceptions are counted as active for audit sampling.", "Cedar release management memo 125: The payment hold is capped at 10 requests during Q3, when the customer tier is enterprise. The note also says archived exceptions remain active.", "Cedar release management memo 125: The payment hold target is at least 12 requests during Q3, when the customer tier is enterprise. Teams may exceed it when capacity is low."], "id": "meaning-q-ed1d73e5895eb6ff", "positive_doc_id": "meaning-doc-27fe8971dfa22ee8", "positive_text": "Cedar release management memo 125: The payment hold is capped at 12 requests during Q3, when the customer tier is enterprise. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Cedar's release management payment hold has a 12 requests cap in Q3 when the customer tier is enterprise, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "48 tickets; only when the request is marked external; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-f0426993345f303b", "meaning-doc-076ddfb892f519d8", "meaning-doc-f9d8bb87833299af"], "hard_negative_texts": ["Aurora clinical triage memo 126: The review level is capped at 48 tickets during Q4, for every request regardless of source. Archived exceptions are counted as active for audit sampling.", "Aurora clinical triage memo 126: The review level is capped at 51 tickets during Q4, only when the request is marked external. The note also says archived exceptions remain active.", "Aurora clinical triage memo 126: The review level target is at least 48 tickets during Q4, only when the request is marked external. Teams may exceed it when capacity is low."], "id": "meaning-q-57c96185c55d506c", "positive_doc_id": "meaning-doc-f3071fc1f77f36cc", "positive_text": "Aurora clinical triage memo 126: The review level is capped at 48 tickets during Q4, only when the request is marked external. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Aurora's clinical triage review level has a 48 tickets cap in Q4 only when the request is marked external, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "18 requests; unless the owner is the platform team; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-a70ca0f756fbd277", "meaning-doc-476decf56daf65c9", "meaning-doc-a9081cc54ca6461d"], "hard_negative_texts": ["Aurora edge caching memo 127: The review level is capped at 18 requests during Q2, for every request regardless of source. Archived exceptions are counted as active for audit sampling.", "Aurora edge caching memo 127: The review level is capped at 15 requests during Q2, unless the owner is the platform team. The note also says archived exceptions remain active.", "Aurora edge caching memo 127: The review level target is at least 18 requests during Q2, unless the owner is the platform team. Teams may exceed it when capacity is low."], "id": "meaning-q-066a0d8aa57e88bf", "positive_doc_id": "meaning-doc-5bb660c6ab369802", "positive_text": "Aurora edge caching memo 127: The review level is capped at 18 requests during Q2, unless the owner is the platform team. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Aurora's edge caching review level has a 18 requests cap in Q2 unless the owner is the platform team, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "7 requests; when the customer tier is enterprise; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-95252b2473c39ba9", "meaning-doc-53c97a494556f5f7", "meaning-doc-6a9783269e437207"], "hard_negative_texts": ["Fjord incident response memo 128: The error budget is capped at 7 requests during Q2, for every request regardless of source. Archived exceptions are counted as active for audit sampling.", "Fjord incident response memo 128: The error budget is capped at 9 requests during Q2, when the customer tier is enterprise. The note also says archived exceptions remain active.", "Fjord incident response memo 128: The error budget target is at least 7 requests during Q2, when the customer tier is enterprise. Teams may exceed it when capacity is low."], "id": "meaning-q-ca3dfb182ccc7b4c", "positive_doc_id": "meaning-doc-5d5b0b27bacddaaa", "positive_text": "Fjord incident response memo 128: The error budget is capped at 7 requests during Q2, when the customer tier is enterprise. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Fjord's incident response error budget has a 7 requests cap in Q2 when the customer tier is enterprise, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "7 hours; when the customer tier is enterprise; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-25671e7d73d3a065", "meaning-doc-e533ba339ef9e279", "meaning-doc-8417db866848621e"], "hard_negative_texts": ["Kepler warehouse operations memo 129: The sampling rate is capped at 7 hours during Q4, after the regional handoff is signed. Archived exceptions are counted as active for audit sampling.", "Kepler warehouse operations memo 129: The sampling rate is capped at 4 hours during Q4, when the customer tier is enterprise. The note also says archived exceptions remain active.", "Kepler warehouse operations memo 129: The sampling rate target is at least 7 hours during Q4, when the customer tier is enterprise. Teams may exceed it when capacity is low."], "id": "meaning-q-ae91549a7d7fc22f", "positive_doc_id": "meaning-doc-ccee3febdb678ac0", "positive_text": "Kepler warehouse operations memo 129: The sampling rate is capped at 7 hours during Q4, when the customer tier is enterprise. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Kepler's warehouse operations sampling rate has a 7 hours cap in Q4 when the customer tier is enterprise, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "24 percent; after the second failed retry; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-c37b068ae67767dd", "meaning-doc-c23a63f0cd03611e", "meaning-doc-9e0cf47684f712a8"], "hard_negative_texts": ["Indigo incident response memo 130: The payment hold is capped at 24 percent during Q3, for every request regardless of source. Archived exceptions are counted as active for audit sampling.", "Indigo incident response memo 130: The payment hold is capped at 30 percent during Q3, after the second failed retry. The note also says archived exceptions remain active.", "Indigo incident response memo 130: The payment hold target is at least 24 percent during Q3, after the second failed retry. Teams may exceed it when capacity is low."], "id": "meaning-q-32d1ed30a8d785d6", "positive_doc_id": "meaning-doc-8f6e74cb8d29a3d3", "positive_text": "Indigo incident response memo 130: The payment hold is capped at 24 percent during Q3, after the second failed retry. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Indigo's incident response payment hold has a 24 percent cap in Q3 after the second failed retry, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "7 hours; before the regional handoff is signed; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-4a4c476d1e3ab152", "meaning-doc-e3d2839f31b459db", "meaning-doc-5af6a9578dbc1cf6"], "hard_negative_texts": ["Fjord payments memo 131: The latency is capped at 7 hours during Q2, for every request regardless of source. Archived exceptions are counted as active for audit sampling.", "Fjord payments memo 131: The latency is capped at 9 hours during Q2, before the regional handoff is signed. The note also says archived exceptions remain active.", "Fjord payments memo 131: The latency target is at least 7 hours during Q2, before the regional handoff is signed. Teams may exceed it when capacity is low."], "id": "meaning-q-1ef87265775e1294", "positive_doc_id": "meaning-doc-6fac4d3b8af9b8d1", "positive_text": "Fjord payments memo 131: The latency is capped at 7 hours during Q2, before the regional handoff is signed. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Fjord's payments latency has a 7 hours cap in Q2 before the regional handoff is signed, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "7 requests; when the customer tier is enterprise; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-73b8c436cc7eaa9d", "meaning-doc-43bd323fe90a7f23", "meaning-doc-ac2feb5a9200f152"], "hard_negative_texts": ["Juniper cloud billing memo 132: The retention period is capped at 7 requests during Q4, when the customer tier is starter. Archived exceptions are counted as active for audit sampling.", "Juniper cloud billing memo 132: The retention period is capped at 5 requests during Q4, when the customer tier is enterprise. The note also says archived exceptions remain active.", "Juniper cloud billing memo 132: The retention period target is at least 7 requests during Q4, when the customer tier is enterprise. Teams may exceed it when capacity is low."], "id": "meaning-q-fbb31d6829535f33", "positive_doc_id": "meaning-doc-82c67a9126c665c4", "positive_text": "Juniper cloud billing memo 132: The retention period is capped at 7 requests during Q4, when the customer tier is enterprise. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Juniper's cloud billing retention period has a 7 requests cap in Q4 when the customer tier is enterprise, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "72 days; when the customer tier is enterprise; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-5f7f22ed5acf0366", "meaning-doc-f5092ba462e2cd55", "meaning-doc-50f595d6f82c41e7"], "hard_negative_texts": ["Basil release management memo 133: The approval threshold is capped at 72 days during Q3, for every request regardless of source. Archived exceptions are counted as active for audit sampling.", "Basil release management memo 133: The approval threshold is capped at 78 days during Q3, when the customer tier is enterprise. The note also says archived exceptions remain active.", "Basil release management memo 133: The approval threshold target is at least 72 days during Q3, when the customer tier is enterprise. Teams may exceed it when capacity is low."], "id": "meaning-q-7823047991ea9b27", "positive_doc_id": "meaning-doc-c23fd62b2b6cee52", "positive_text": "Basil release management memo 133: The approval threshold is capped at 72 days during Q3, when the customer tier is enterprise. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Basil's release management approval threshold has a 72 days cap in Q3 when the customer tier is enterprise, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "12 days; when the customer tier is enterprise; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-742340d1cb0d9248", "meaning-doc-6639e493e9cc149c", "meaning-doc-1dc6c475b538bd3a"], "hard_negative_texts": ["Juniper privacy requests memo 134: The cache lifetime is capped at 12 days during Q4, only when the owner is the platform team. Archived exceptions are counted as active for audit sampling.", "Juniper privacy requests memo 134: The cache lifetime is capped at 15 days during Q4, when the customer tier is enterprise. The note also says archived exceptions remain active.", "Juniper privacy requests memo 134: The cache lifetime target is at least 12 days during Q4, when the customer tier is enterprise. Teams may exceed it when capacity is low."], "id": "meaning-q-0f7a314d91006576", "positive_doc_id": "meaning-doc-f3aa2ccd84fd238c", "positive_text": "Juniper privacy requests memo 134: The cache lifetime is capped at 12 days during Q4, when the customer tier is enterprise. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Juniper's privacy requests cache lifetime has a 12 days cap in Q4 when the customer tier is enterprise, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "72 minutes; before the regional handoff is signed; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-f94cb06984e5d92a", "meaning-doc-f64eaf707b97f4e2", "meaning-doc-f77eb4a6bee42ec8"], "hard_negative_texts": ["Kepler warehouse operations memo 135: The review level is capped at 72 minutes during Q4, when the customer tier is starter. Archived exceptions are counted as active for audit sampling.", "Kepler warehouse operations memo 135: The review level is capped at 70 minutes during Q4, before the regional handoff is signed. The note also says archived exceptions remain active.", "Kepler warehouse operations memo 135: The review level target is at least 72 minutes during Q4, before the regional handoff is signed. Teams may exceed it when capacity is low."], "id": "meaning-q-7472d51b464ced0a", "positive_doc_id": "meaning-doc-7ac1e05ea090e19e", "positive_text": "Kepler warehouse operations memo 135: The review level is capped at 72 minutes during Q4, before the regional handoff is signed. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Kepler's warehouse operations review level has a 72 minutes cap in Q4 before the regional handoff is signed, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "48 minutes; only when the request is marked external; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-0e3266360776201e", "meaning-doc-8765356feded5391", "meaning-doc-a8e8617bb21cae86"], "hard_negative_texts": ["Fjord incident response memo 136: The escalation limit is capped at 48 minutes during Q1, for every request regardless of source. Archived exceptions are counted as active for audit sampling.", "Fjord incident response memo 136: The escalation limit is capped at 45 minutes during Q1, only when the request is marked external. The note also says archived exceptions remain active.", "Fjord incident response memo 136: The escalation limit target is at least 48 minutes during Q1, only when the request is marked external. Teams may exceed it when capacity is low."], "id": "meaning-q-9f99ed745cacd642", "positive_doc_id": "meaning-doc-1d4ea77fbbf85e0a", "positive_text": "Fjord incident response memo 136: The escalation limit is capped at 48 minutes during Q1, only when the request is marked external. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Fjord's incident response escalation limit has a 48 minutes cap in Q1 only when the request is marked external, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "48 percent; after the second failed retry; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-678e4f4d08a146a0", "meaning-doc-081b8103b3999b8d", "meaning-doc-4d9db1ecbcebf25b"], "hard_negative_texts": ["Cedar model deployment memo 137: The rollback window is capped at 48 percent during Q2, after the first successful retry. Archived exceptions are counted as active for audit sampling.", "Cedar model deployment memo 137: The rollback window is capped at 54 percent during Q2, after the second failed retry. The note also says archived exceptions remain active.", "Cedar model deployment memo 137: The rollback window target is at least 48 percent during Q2, after the second failed retry. Teams may exceed it when capacity is low."], "id": "meaning-q-60cf14f2c1ccdfe4", "positive_doc_id": "meaning-doc-3ed5659ec6514286", "positive_text": "Cedar model deployment memo 137: The rollback window is capped at 48 percent during Q2, after the second failed retry. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Cedar's model deployment rollback window has a 48 percent cap in Q2 after the second failed retry, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "18 requests; after the second failed retry; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-07589ac160df4dd7", "meaning-doc-0951b5a29fa670a2", "meaning-doc-bf050cc245a2103b"], "hard_negative_texts": ["Fjord payments memo 138: The rollback window is capped at 18 requests during Q2, after the first successful retry. Archived exceptions are counted as active for audit sampling.", "Fjord payments memo 138: The rollback window is capped at 24 requests during Q2, after the second failed retry. The note also says archived exceptions remain active.", "Fjord payments memo 138: The rollback window target is at least 18 requests during Q2, after the second failed retry. Teams may exceed it when capacity is low."], "id": "meaning-q-ca0f4c62e2c06324", "positive_doc_id": "meaning-doc-f6a3dd19986b36ef", "positive_text": "Fjord payments memo 138: The rollback window is capped at 18 requests during Q2, after the second failed retry. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Fjord's payments rollback window has a 18 requests cap in Q2 after the second failed retry, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "12 requests; before the regional handoff is signed; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-dd3c711b0c541cf1", "meaning-doc-7939d9c0836f7681", "meaning-doc-9b32782f57815822"], "hard_negative_texts": ["Lumen privacy requests memo 139: The escalation limit is capped at 12 requests during Q3, for every request regardless of source. Archived exceptions are counted as active for audit sampling.", "Lumen privacy requests memo 139: The escalation limit is capped at 10 requests during Q3, before the regional handoff is signed. The note also says archived exceptions remain active.", "Lumen privacy requests memo 139: The escalation limit target is at least 12 requests during Q3, before the regional handoff is signed. Teams may exceed it when capacity is low."], "id": "meaning-q-bcaaaf3e69cc7ebc", "positive_doc_id": "meaning-doc-4b451d0ecf32f03e", "positive_text": "Lumen privacy requests memo 139: The escalation limit is capped at 12 requests during Q3, before the regional handoff is signed. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Lumen's privacy requests escalation limit has a 12 requests cap in Q3 before the regional handoff is signed, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "24 hours; unless the owner is the platform team; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-206af19d0667585a", "meaning-doc-94e1e39e11e775e7", "meaning-doc-39fb72ab3d31470b"], "hard_negative_texts": ["Ember payments memo 140: The escalation limit is capped at 24 hours during Q1, only when the owner is the platform team. Archived exceptions are counted as active for audit sampling.", "Ember payments memo 140: The escalation limit is capped at 21 hours during Q1, unless the owner is the platform team. The note also says archived exceptions remain active.", "Ember payments memo 140: The escalation limit target is at least 24 hours during Q1, unless the owner is the platform team. Teams may exceed it when capacity is low."], "id": "meaning-q-0abdf21dc7f98af8", "positive_doc_id": "meaning-doc-f2f87b32942e63f4", "positive_text": "Ember payments memo 140: The escalation limit is capped at 24 hours during Q1, unless the owner is the platform team. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Ember's payments escalation limit has a 24 hours cap in Q1 unless the owner is the platform team, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "7 hours; when the customer tier is enterprise; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-2c54c324d2649d81", "meaning-doc-0d0fab620f5de59d", "meaning-doc-0ba34b427815daae"], "hard_negative_texts": ["Fjord privacy requests memo 141: The rollback window is capped at 7 hours during Q3, for every request regardless of source. Archived exceptions are counted as active for audit sampling.", "Fjord privacy requests memo 141: The rollback window is capped at 13 hours during Q3, when the customer tier is enterprise. The note also says archived exceptions remain active.", "Fjord privacy requests memo 141: The rollback window target is at least 7 hours during Q3, when the customer tier is enterprise. Teams may exceed it when capacity is low."], "id": "meaning-q-7ed1df77d947cdcd", "positive_doc_id": "meaning-doc-e9439d8369170f97", "positive_text": "Fjord privacy requests memo 141: The rollback window is capped at 7 hours during Q3, when the customer tier is enterprise. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Fjord's privacy requests rollback window has a 7 hours cap in Q3 when the customer tier is enterprise, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "96 tickets; after the second failed retry; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-64ab077d297f61be", "meaning-doc-3d0736d13b67c45a", "meaning-doc-fdaf1789d2068788"], "hard_negative_texts": ["Basil model deployment memo 142: The retention period is capped at 96 tickets during Q3, after the regional handoff is signed. Archived exceptions are counted as active for audit sampling.", "Basil model deployment memo 142: The retention period is capped at 102 tickets during Q3, after the second failed retry. The note also says archived exceptions remain active.", "Basil model deployment memo 142: The retention period target is at least 96 tickets during Q3, after the second failed retry. Teams may exceed it when capacity is low."], "id": "meaning-q-9755f557a420f324", "positive_doc_id": "meaning-doc-2697981fe7003cc0", "positive_text": "Basil model deployment memo 142: The retention period is capped at 96 tickets during Q3, after the second failed retry. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Basil's model deployment retention period has a 96 tickets cap in Q3 after the second failed retry, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "96 tickets; before the regional handoff is signed; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-abc84d9a5c0d1b3b", "meaning-doc-6e3a3551638a8f2f", "meaning-doc-4bd8ceb594f63270"], "hard_negative_texts": ["Fjord cloud billing memo 143: The latency is capped at 96 tickets during Q1, after the regional handoff is signed. Archived exceptions are counted as active for audit sampling.", "Fjord cloud billing memo 143: The latency is capped at 99 tickets during Q1, before the regional handoff is signed. The note also says archived exceptions remain active.", "Fjord cloud billing memo 143: The latency target is at least 96 tickets during Q1, before the regional handoff is signed. Teams may exceed it when capacity is low."], "id": "meaning-q-cd385c9852ef2ea5", "positive_doc_id": "meaning-doc-dfd9f59cc1acaaec", "positive_text": "Fjord cloud billing memo 143: The latency is capped at 96 tickets during Q1, before the regional handoff is signed. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Fjord's cloud billing latency has a 96 tickets cap in Q1 before the regional handoff is signed, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "12 hours; only when the request is marked external; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-62df42a79825e094", "meaning-doc-95513c105f0c93ef", "meaning-doc-dd8d51e7ebf656b9"], "hard_negative_texts": ["Fjord warehouse operations memo 144: The cache lifetime is capped at 12 hours during Q4, when the customer tier is starter. Archived exceptions are counted as active for audit sampling.", "Fjord warehouse operations memo 144: The cache lifetime is capped at 18 hours during Q4, only when the request is marked external. The note also says archived exceptions remain active.", "Fjord warehouse operations memo 144: The cache lifetime target is at least 12 hours during Q4, only when the request is marked external. Teams may exceed it when capacity is low."], "id": "meaning-q-20caed4ae7e99361", "positive_doc_id": "meaning-doc-a4d83550e458bdf6", "positive_text": "Fjord warehouse operations memo 144: The cache lifetime is capped at 12 hours during Q4, only when the request is marked external. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Fjord's warehouse operations cache lifetime has a 12 hours cap in Q4 only when the request is marked external, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "96 tickets; before the regional handoff is signed; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-94107b6f616e56d1", "meaning-doc-4eefc68cb2b8997b", "meaning-doc-d3abb023477ea2dc"], "hard_negative_texts": ["Basil privacy requests memo 145: The rollback window is capped at 96 tickets during Q1, only when the owner is the platform team. Archived exceptions are counted as active for audit sampling.", "Basil privacy requests memo 145: The rollback window is capped at 98 tickets during Q1, before the regional handoff is signed. The note also says archived exceptions remain active.", "Basil privacy requests memo 145: The rollback window target is at least 96 tickets during Q1, before the regional handoff is signed. Teams may exceed it when capacity is low."], "id": "meaning-q-8a3a2d001b7893a7", "positive_doc_id": "meaning-doc-43f4a1bdde0836ac", "positive_text": "Basil privacy requests memo 145: The rollback window is capped at 96 tickets during Q1, before the regional handoff is signed. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Basil's privacy requests rollback window has a 96 tickets cap in Q1 before the regional handoff is signed, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "72 percent; unless the owner is the platform team; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-0a3465f11ef55e9f", "meaning-doc-4f519dc2af42c1cb", "meaning-doc-19af82907eb3b374"], "hard_negative_texts": ["Lumen model deployment memo 146: The sampling rate is capped at 72 percent during Q4, after the regional handoff is signed. Archived exceptions are counted as active for audit sampling.", "Lumen model deployment memo 146: The sampling rate is capped at 70 percent during Q4, unless the owner is the platform team. The note also says archived exceptions remain active.", "Lumen model deployment memo 146: The sampling rate target is at least 72 percent during Q4, unless the owner is the platform team. Teams may exceed it when capacity is low."], "id": "meaning-q-9e0ae531a7baafa6", "positive_doc_id": "meaning-doc-07aadd3cad8216b0", "positive_text": "Lumen model deployment memo 146: The sampling rate is capped at 72 percent during Q4, unless the owner is the platform team. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Lumen's model deployment sampling rate has a 72 percent cap in Q4 unless the owner is the platform team, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "7 hours; when the customer tier is enterprise; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-b9c4be9a107761d1", "meaning-doc-ac290ad1f860ae69", "meaning-doc-29f617a36f05dbaa"], "hard_negative_texts": ["Cedar cloud billing memo 147: The cache lifetime is capped at 7 hours during Q4, when the customer tier is starter. Archived exceptions are counted as active for audit sampling.", "Cedar cloud billing memo 147: The cache lifetime is capped at 13 hours during Q4, when the customer tier is enterprise. The note also says archived exceptions remain active.", "Cedar cloud billing memo 147: The cache lifetime target is at least 7 hours during Q4, when the customer tier is enterprise. Teams may exceed it when capacity is low."], "id": "meaning-q-3d32d5f3c252605d", "positive_doc_id": "meaning-doc-7e98040f3622fa74", "positive_text": "Cedar cloud billing memo 147: The cache lifetime is capped at 7 hours during Q4, when the customer tier is enterprise. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Cedar's cloud billing cache lifetime has a 7 hours cap in Q4 when the customer tier is enterprise, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "12 hours; unless the owner is the platform team; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-ccef1fb9ca7ba060", "meaning-doc-2ce9d23182f4d70f", "meaning-doc-95431d22ca34e54a"], "hard_negative_texts": ["Juniper incident response memo 148: The retention period is capped at 12 hours during Q4, when the customer tier is starter. Archived exceptions are counted as active for audit sampling.", "Juniper incident response memo 148: The retention period is capped at 14 hours during Q4, unless the owner is the platform team. The note also says archived exceptions remain active.", "Juniper incident response memo 148: The retention period target is at least 12 hours during Q4, unless the owner is the platform team. Teams may exceed it when capacity is low."], "id": "meaning-q-7077d37b7824ac54", "positive_doc_id": "meaning-doc-e72c085a0ce98511", "positive_text": "Juniper incident response memo 148: The retention period is capped at 12 hours during Q4, unless the owner is the platform team. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Juniper's incident response retention period has a 12 hours cap in Q4 unless the owner is the platform team, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "96 minutes; when the customer tier is enterprise; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-e190708329608af7", "meaning-doc-fd1a45155ff8ef4a", "meaning-doc-4a865ab45fa6facb"], "hard_negative_texts": ["Basil payments memo 149: The review level is capped at 96 minutes during Q1, when the customer tier is starter. Archived exceptions are counted as active for audit sampling.", "Basil payments memo 149: The review level is capped at 94 minutes during Q1, when the customer tier is enterprise. The note also says archived exceptions remain active.", "Basil payments memo 149: The review level target is at least 96 minutes during Q1, when the customer tier is enterprise. Teams may exceed it when capacity is low."], "id": "meaning-q-45c69f1cf3880a3d", "positive_doc_id": "meaning-doc-1e2906ccb19061c0", "positive_text": "Basil payments memo 149: The review level is capped at 96 minutes during Q1, when the customer tier is enterprise. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Basil's payments review level has a 96 minutes cap in Q1 when the customer tier is enterprise, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "96 percent; before the regional handoff is signed; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-2830297e95094185", "meaning-doc-5cbab0f8ebe71330", "meaning-doc-966f92e6866a84f5"], "hard_negative_texts": ["Cedar data retention memo 150: The rollback window is capped at 96 percent during Q3, for every request regardless of source. Archived exceptions are counted as active for audit sampling.", "Cedar data retention memo 150: The rollback window is capped at 98 percent during Q3, before the regional handoff is signed. The note also says archived exceptions remain active.", "Cedar data retention memo 150: The rollback window target is at least 96 percent during Q3, before the regional handoff is signed. Teams may exceed it when capacity is low."], "id": "meaning-q-5320bc7a4ff533d1", "positive_doc_id": "meaning-doc-d3de950360364388", "positive_text": "Cedar data retention memo 150: The rollback window is capped at 96 percent during Q3, before the regional handoff is signed. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Cedar's data retention rollback window has a 96 percent cap in Q3 before the regional handoff is signed, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "48 tickets; unless the owner is the platform team; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-047b45075ddbd809", "meaning-doc-b6492d1cc7409f52", "meaning-doc-9e3f4cdef387aa92"], "hard_negative_texts": ["Aurora warehouse operations memo 151: The review level is capped at 48 tickets during Q2, when the customer tier is starter. Archived exceptions are counted as active for audit sampling.", "Aurora warehouse operations memo 151: The review level is capped at 54 tickets during Q2, unless the owner is the platform team. The note also says archived exceptions remain active.", "Aurora warehouse operations memo 151: The review level target is at least 48 tickets during Q2, unless the owner is the platform team. Teams may exceed it when capacity is low."], "id": "meaning-q-8cf553d651584fc4", "positive_doc_id": "meaning-doc-aa809f6387e14669", "positive_text": "Aurora warehouse operations memo 151: The review level is capped at 48 tickets during Q2, unless the owner is the platform team. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Aurora's warehouse operations review level has a 48 tickets cap in Q2 unless the owner is the platform team, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "24 percent; after the second failed retry; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-e44ac3620908f10b", "meaning-doc-f112c75963bfcc16", "meaning-doc-b75d6d178d848217"], "hard_negative_texts": ["Harbor warehouse operations memo 152: The latency is capped at 24 percent during Q3, after the first successful retry. Archived exceptions are counted as active for audit sampling.", "Harbor warehouse operations memo 152: The latency is capped at 30 percent during Q3, after the second failed retry. The note also says archived exceptions remain active.", "Harbor warehouse operations memo 152: The latency target is at least 24 percent during Q3, after the second failed retry. Teams may exceed it when capacity is low."], "id": "meaning-q-c2ed9338ea90813b", "positive_doc_id": "meaning-doc-d39c4513e10f0d12", "positive_text": "Harbor warehouse operations memo 152: The latency is capped at 24 percent during Q3, after the second failed retry. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Harbor's warehouse operations latency has a 24 percent cap in Q3 after the second failed retry, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "7 days; before the regional handoff is signed; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-e56c5fcc4898d278", "meaning-doc-62232be8b2da4c10", "meaning-doc-f4a62d67ba92d60b"], "hard_negative_texts": ["Lumen security review memo 153: The escalation limit is capped at 7 days during Q4, after the regional handoff is signed. Archived exceptions are counted as active for audit sampling.", "Lumen security review memo 153: The escalation limit is capped at 10 days during Q4, before the regional handoff is signed. The note also says archived exceptions remain active.", "Lumen security review memo 153: The escalation limit target is at least 7 days during Q4, before the regional handoff is signed. Teams may exceed it when capacity is low."], "id": "meaning-q-6945700d3adc9330", "positive_doc_id": "meaning-doc-865897f634beced7", "positive_text": "Lumen security review memo 153: The escalation limit is capped at 7 days during Q4, before the regional handoff is signed. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Lumen's security review escalation limit has a 7 days cap in Q4 before the regional handoff is signed, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "24 tickets; only when the request is marked external; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-681bbe74a559fdfd", "meaning-doc-05096fff7a58467c", "meaning-doc-787882b8aa7a16cd"], "hard_negative_texts": ["Fjord security review memo 154: The error budget is capped at 24 tickets during Q4, when the customer tier is starter. Archived exceptions are counted as active for audit sampling.", "Fjord security review memo 154: The error budget is capped at 27 tickets during Q4, only when the request is marked external. The note also says archived exceptions remain active.", "Fjord security review memo 154: The error budget target is at least 24 tickets during Q4, only when the request is marked external. Teams may exceed it when capacity is low."], "id": "meaning-q-4adffaea03f14563", "positive_doc_id": "meaning-doc-963458c134df9aaf", "positive_text": "Fjord security review memo 154: The error budget is capped at 24 tickets during Q4, only when the request is marked external. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Fjord's security review error budget has a 24 tickets cap in Q4 only when the request is marked external, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "96 minutes; only when the request is marked external; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-6a3a05f865fcd063", "meaning-doc-160c2ed08628c705", "meaning-doc-169fc13b6786dd2d"], "hard_negative_texts": ["Garnet warehouse operations memo 155: The rollback window is capped at 96 minutes during Q4, only when the owner is the platform team. Archived exceptions are counted as active for audit sampling.", "Garnet warehouse operations memo 155: The rollback window is capped at 102 minutes during Q4, only when the request is marked external. The note also says archived exceptions remain active.", "Garnet warehouse operations memo 155: The rollback window target is at least 96 minutes during Q4, only when the request is marked external. Teams may exceed it when capacity is low."], "id": "meaning-q-2b16ceb7df006f01", "positive_doc_id": "meaning-doc-90038e0711f51986", "positive_text": "Garnet warehouse operations memo 155: The rollback window is capped at 96 minutes during Q4, only when the request is marked external. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Garnet's warehouse operations rollback window has a 96 minutes cap in Q4 only when the request is marked external, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "12 percent; before the regional handoff is signed; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-c44883ff87a18a10", "meaning-doc-de32b2e896051fec", "meaning-doc-64424dda37851b00"], "hard_negative_texts": ["Cedar procurement memo 156: The escalation limit is capped at 12 percent during Q1, only when the owner is the platform team. Archived exceptions are counted as active for audit sampling.", "Cedar procurement memo 156: The escalation limit is capped at 18 percent during Q1, before the regional handoff is signed. The note also says archived exceptions remain active.", "Cedar procurement memo 156: The escalation limit target is at least 12 percent during Q1, before the regional handoff is signed. Teams may exceed it when capacity is low."], "id": "meaning-q-5a3e06f657726505", "positive_doc_id": "meaning-doc-1b1a295667d20dcc", "positive_text": "Cedar procurement memo 156: The escalation limit is capped at 12 percent during Q1, before the regional handoff is signed. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Cedar's procurement escalation limit has a 12 percent cap in Q1 before the regional handoff is signed, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "12 minutes; after the second failed retry; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-fff0400f1c5d6701", "meaning-doc-7b2dcf3701552c3b", "meaning-doc-5b5ee47e03408f3c"], "hard_negative_texts": ["Garnet procurement memo 157: The cache lifetime is capped at 12 minutes during Q4, for every request regardless of source. Archived exceptions are counted as active for audit sampling.", "Garnet procurement memo 157: The cache lifetime is capped at 15 minutes during Q4, after the second failed retry. The note also says archived exceptions remain active.", "Garnet procurement memo 157: The cache lifetime target is at least 12 minutes during Q4, after the second failed retry. Teams may exceed it when capacity is low."], "id": "meaning-q-8038aa3c694956a1", "positive_doc_id": "meaning-doc-ba3233538db476fe", "positive_text": "Garnet procurement memo 157: The cache lifetime is capped at 12 minutes during Q4, after the second failed retry. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Garnet's procurement cache lifetime has a 12 minutes cap in Q4 after the second failed retry, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "7 requests; before the regional handoff is signed; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-5a9f3d0811794fe0", "meaning-doc-9d6ced6f5441f270", "meaning-doc-a72f204d13de6c4d"], "hard_negative_texts": ["Harbor clinical triage memo 158: The review level is capped at 7 requests during Q2, when the customer tier is starter. Archived exceptions are counted as active for audit sampling.", "Harbor clinical triage memo 158: The review level is capped at 9 requests during Q2, before the regional handoff is signed. The note also says archived exceptions remain active.", "Harbor clinical triage memo 158: The review level target is at least 7 requests during Q2, before the regional handoff is signed. Teams may exceed it when capacity is low."], "id": "meaning-q-40e5ecbec68fb709", "positive_doc_id": "meaning-doc-24b631aa2d93ea47", "positive_text": "Harbor clinical triage memo 158: The review level is capped at 7 requests during Q2, before the regional handoff is signed. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Harbor's clinical triage review level has a 7 requests cap in Q2 before the regional handoff is signed, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "24 days; after the second failed retry; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-600466571b269b50", "meaning-doc-c3b1dddd2087be0a", "meaning-doc-ce145cc05472784e"], "hard_negative_texts": ["Cedar release management memo 159: The cache lifetime is capped at 24 days during Q1, for every request regardless of source. Archived exceptions are counted as active for audit sampling.", "Cedar release management memo 159: The cache lifetime is capped at 26 days during Q1, after the second failed retry. The note also says archived exceptions remain active.", "Cedar release management memo 159: The cache lifetime target is at least 24 days during Q1, after the second failed retry. Teams may exceed it when capacity is low."], "id": "meaning-q-a0e52ca5b9130b50", "positive_doc_id": "meaning-doc-20ca81f0b4113aec", "positive_text": "Cedar release management memo 159: The cache lifetime is capped at 24 days during Q1, after the second failed retry. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Cedar's release management cache lifetime has a 24 days cap in Q1 after the second failed retry, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "72 hours; before the regional handoff is signed; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-4fe9c7d6f70f5d82", "meaning-doc-83ba289823f15837", "meaning-doc-a4c7b84ed275c40c"], "hard_negative_texts": ["Basil release management memo 160: The review level is capped at 72 hours during Q1, when the customer tier is starter. Archived exceptions are counted as active for audit sampling.", "Basil release management memo 160: The review level is capped at 69 hours during Q1, before the regional handoff is signed. The note also says archived exceptions remain active.", "Basil release management memo 160: The review level target is at least 72 hours during Q1, before the regional handoff is signed. Teams may exceed it when capacity is low."], "id": "meaning-q-8b5183c68a4ca985", "positive_doc_id": "meaning-doc-6175978788f28c68", "positive_text": "Basil release management memo 160: The review level is capped at 72 hours during Q1, before the regional handoff is signed. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Basil's release management review level has a 72 hours cap in Q1 before the regional handoff is signed, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "36 days; unless the owner is the platform team; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-3908d079310a92c6", "meaning-doc-0e47998f9d2eeb11", "meaning-doc-ea48e9ff64393842"], "hard_negative_texts": ["Indigo clinical triage memo 161: The cache lifetime is capped at 36 days during Q1, for every request regardless of source. Archived exceptions are counted as active for audit sampling.", "Indigo clinical triage memo 161: The cache lifetime is capped at 42 days during Q1, unless the owner is the platform team. The note also says archived exceptions remain active.", "Indigo clinical triage memo 161: The cache lifetime target is at least 36 days during Q1, unless the owner is the platform team. Teams may exceed it when capacity is low."], "id": "meaning-q-e22566ea0d6ca3b6", "positive_doc_id": "meaning-doc-f29a388ab26cd8db", "positive_text": "Indigo clinical triage memo 161: The cache lifetime is capped at 36 days during Q1, unless the owner is the platform team. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Indigo's clinical triage cache lifetime has a 36 days cap in Q1 unless the owner is the platform team, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "48 percent; after the second failed retry; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-0ac8b18e57576cc0", "meaning-doc-bc0fcd64602f6d23", "meaning-doc-402d4a800c4bbc1b"], "hard_negative_texts": ["Juniper payments memo 162: The cache lifetime is capped at 48 percent during Q4, after the first successful retry. Archived exceptions are counted as active for audit sampling.", "Juniper payments memo 162: The cache lifetime is capped at 54 percent during Q4, after the second failed retry. The note also says archived exceptions remain active.", "Juniper payments memo 162: The cache lifetime target is at least 48 percent during Q4, after the second failed retry. Teams may exceed it when capacity is low."], "id": "meaning-q-920294f2802dc787", "positive_doc_id": "meaning-doc-ad903a3d96a0cd2b", "positive_text": "Juniper payments memo 162: The cache lifetime is capped at 48 percent during Q4, after the second failed retry. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Juniper's payments cache lifetime has a 48 percent cap in Q4 after the second failed retry, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "24 tickets; when the customer tier is enterprise; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-cf86b0cdb95d90b7", "meaning-doc-0836902babd6fd7f", "meaning-doc-56c1f43f996844e0"], "hard_negative_texts": ["Fjord release management memo 163: The review level is capped at 24 tickets during Q4, only when the owner is the platform team. Archived exceptions are counted as active for audit sampling.", "Fjord release management memo 163: The review level is capped at 22 tickets during Q4, when the customer tier is enterprise. The note also says archived exceptions remain active.", "Fjord release management memo 163: The review level target is at least 24 tickets during Q4, when the customer tier is enterprise. Teams may exceed it when capacity is low."], "id": "meaning-q-ec99b284f4dcbc5a", "positive_doc_id": "meaning-doc-4c8baeaea1b9cedd", "positive_text": "Fjord release management memo 163: The review level is capped at 24 tickets during Q4, when the customer tier is enterprise. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Fjord's release management review level has a 24 tickets cap in Q4 when the customer tier is enterprise, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "72 percent; when the customer tier is enterprise; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-668237567d6bc588", "meaning-doc-25f00e5adf77913d", "meaning-doc-adc1aaf067612d7c"], "hard_negative_texts": ["Fjord incident response memo 164: The sampling rate is capped at 72 percent during Q2, after the first successful retry. Archived exceptions are counted as active for audit sampling.", "Fjord incident response memo 164: The sampling rate is capped at 70 percent during Q2, when the customer tier is enterprise. The note also says archived exceptions remain active.", "Fjord incident response memo 164: The sampling rate target is at least 72 percent during Q2, when the customer tier is enterprise. Teams may exceed it when capacity is low."], "id": "meaning-q-e477ecf3ff823e19", "positive_doc_id": "meaning-doc-5e4b00783360fa17", "positive_text": "Fjord incident response memo 164: The sampling rate is capped at 72 percent during Q2, when the customer tier is enterprise. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Fjord's incident response sampling rate has a 72 percent cap in Q2 when the customer tier is enterprise, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "12 requests; unless the owner is the platform team; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-6afe9080195dc988", "meaning-doc-3cbfc56224827608", "meaning-doc-d650bfcf1d124617"], "hard_negative_texts": ["Cedar data retention memo 165: The rollback window is capped at 12 requests during Q1, after the first successful retry. Archived exceptions are counted as active for audit sampling.", "Cedar data retention memo 165: The rollback window is capped at 10 requests during Q1, unless the owner is the platform team. The note also says archived exceptions remain active.", "Cedar data retention memo 165: The rollback window target is at least 12 requests during Q1, unless the owner is the platform team. Teams may exceed it when capacity is low."], "id": "meaning-q-e26abf9ebfd7ff59", "positive_doc_id": "meaning-doc-53abf1751a8818e7", "positive_text": "Cedar data retention memo 165: The rollback window is capped at 12 requests during Q1, unless the owner is the platform team. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Cedar's data retention rollback window has a 12 requests cap in Q1 unless the owner is the platform team, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "24 hours; unless the owner is the platform team; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-9d5799ab2d865e24", "meaning-doc-204951ab0629bd2c", "meaning-doc-e6aeb91575ac4c99"], "hard_negative_texts": ["Basil release management memo 166: The payment hold is capped at 24 hours during Q3, after the regional handoff is signed. Archived exceptions are counted as active for audit sampling.", "Basil release management memo 166: The payment hold is capped at 21 hours during Q3, unless the owner is the platform team. The note also says archived exceptions remain active.", "Basil release management memo 166: The payment hold target is at least 24 hours during Q3, unless the owner is the platform team. Teams may exceed it when capacity is low."], "id": "meaning-q-f0d34e6884af9d21", "positive_doc_id": "meaning-doc-0dceb7735aea6eff", "positive_text": "Basil release management memo 166: The payment hold is capped at 24 hours during Q3, unless the owner is the platform team. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Basil's release management payment hold has a 24 hours cap in Q3 unless the owner is the platform team, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "72 requests; after the second failed retry; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-34c62854b38b8151", "meaning-doc-d366e7157c8c6289", "meaning-doc-c3a878d7f9494123"], "hard_negative_texts": ["Lumen release management memo 167: The approval threshold is capped at 72 requests during Q1, after the first successful retry. Archived exceptions are counted as active for audit sampling.", "Lumen release management memo 167: The approval threshold is capped at 78 requests during Q1, after the second failed retry. The note also says archived exceptions remain active.", "Lumen release management memo 167: The approval threshold target is at least 72 requests during Q1, after the second failed retry. Teams may exceed it when capacity is low."], "id": "meaning-q-cca3f011a45869d3", "positive_doc_id": "meaning-doc-3a07620b7c58209b", "positive_text": "Lumen release management memo 167: The approval threshold is capped at 72 requests during Q1, after the second failed retry. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Lumen's release management approval threshold has a 72 requests cap in Q1 after the second failed retry, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "48 minutes; when the customer tier is enterprise; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-1a97798cdb0eb37f", "meaning-doc-71dae65c4dc09ea3", "meaning-doc-39c0ce4e2f675e99"], "hard_negative_texts": ["Lumen privacy requests memo 168: The payment hold is capped at 48 minutes during Q2, for every request regardless of source. Archived exceptions are counted as active for audit sampling.", "Lumen privacy requests memo 168: The payment hold is capped at 50 minutes during Q2, when the customer tier is enterprise. The note also says archived exceptions remain active.", "Lumen privacy requests memo 168: The payment hold target is at least 48 minutes during Q2, when the customer tier is enterprise. Teams may exceed it when capacity is low."], "id": "meaning-q-f62c876fd46e6251", "positive_doc_id": "meaning-doc-49ff5c71c9d74c0d", "positive_text": "Lumen privacy requests memo 168: The payment hold is capped at 48 minutes during Q2, when the customer tier is enterprise. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Lumen's privacy requests payment hold has a 48 minutes cap in Q2 when the customer tier is enterprise, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "18 minutes; after the second failed retry; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-74f027f8b49d2674", "meaning-doc-37e09db7bd813f0d", "meaning-doc-f3988622ed535141"], "hard_negative_texts": ["Cedar incident response memo 169: The cache lifetime is capped at 18 minutes during Q4, when the customer tier is starter. Archived exceptions are counted as active for audit sampling.", "Cedar incident response memo 169: The cache lifetime is capped at 20 minutes during Q4, after the second failed retry. The note also says archived exceptions remain active.", "Cedar incident response memo 169: The cache lifetime target is at least 18 minutes during Q4, after the second failed retry. Teams may exceed it when capacity is low."], "id": "meaning-q-c71a69f862db38ca", "positive_doc_id": "meaning-doc-a39a77636b74b0fb", "positive_text": "Cedar incident response memo 169: The cache lifetime is capped at 18 minutes during Q4, after the second failed retry. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Cedar's incident response cache lifetime has a 18 minutes cap in Q4 after the second failed retry, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "48 minutes; before the regional handoff is signed; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-d2c87066315b9c41", "meaning-doc-9b86ef073f633992", "meaning-doc-f2d2ca1f1f3c2a9c"], "hard_negative_texts": ["Aurora release management memo 170: The latency is capped at 48 minutes during Q2, after the regional handoff is signed. Archived exceptions are counted as active for audit sampling.", "Aurora release management memo 170: The latency is capped at 54 minutes during Q2, before the regional handoff is signed. The note also says archived exceptions remain active.", "Aurora release management memo 170: The latency target is at least 48 minutes during Q2, before the regional handoff is signed. Teams may exceed it when capacity is low."], "id": "meaning-q-5f90ccc38714ae2d", "positive_doc_id": "meaning-doc-09c458d956d3460d", "positive_text": "Aurora release management memo 170: The latency is capped at 48 minutes during Q2, before the regional handoff is signed. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Aurora's release management latency has a 48 minutes cap in Q2 before the regional handoff is signed, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "96 days; unless the owner is the platform team; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-1e5a8f6053cb6322", "meaning-doc-d4f9dc153bd6d01d", "meaning-doc-770a26ffcda43cec"], "hard_negative_texts": ["Lumen incident response memo 171: The escalation limit is capped at 96 days during Q1, for every request regardless of source. Archived exceptions are counted as active for audit sampling.", "Lumen incident response memo 171: The escalation limit is capped at 99 days during Q1, unless the owner is the platform team. The note also says archived exceptions remain active.", "Lumen incident response memo 171: The escalation limit target is at least 96 days during Q1, unless the owner is the platform team. Teams may exceed it when capacity is low."], "id": "meaning-q-b067f6741f084421", "positive_doc_id": "meaning-doc-097c35a43c9879c0", "positive_text": "Lumen incident response memo 171: The escalation limit is capped at 96 days during Q1, unless the owner is the platform team. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Lumen's incident response escalation limit has a 96 days cap in Q1 unless the owner is the platform team, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "72 days; before the regional handoff is signed; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-7516622e713ba76b", "meaning-doc-4f9d30543d1b8aa8", "meaning-doc-3d222d9ba54581cb"], "hard_negative_texts": ["Lumen warehouse operations memo 172: The sampling rate is capped at 72 days during Q2, when the customer tier is starter. Archived exceptions are counted as active for audit sampling.", "Lumen warehouse operations memo 172: The sampling rate is capped at 78 days during Q2, before the regional handoff is signed. The note also says archived exceptions remain active.", "Lumen warehouse operations memo 172: The sampling rate target is at least 72 days during Q2, before the regional handoff is signed. Teams may exceed it when capacity is low."], "id": "meaning-q-720781cea6ea9e75", "positive_doc_id": "meaning-doc-5c1a10eaa73df928", "positive_text": "Lumen warehouse operations memo 172: The sampling rate is capped at 72 days during Q2, before the regional handoff is signed. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Lumen's warehouse operations sampling rate has a 72 days cap in Q2 before the regional handoff is signed, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "7 days; before the regional handoff is signed; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-bc70472c348e9023", "meaning-doc-5ca60fa4942df73b", "meaning-doc-a1b74cfd9d857726"], "hard_negative_texts": ["Garnet data retention memo 173: The cache lifetime is capped at 7 days during Q1, only when the owner is the platform team. Archived exceptions are counted as active for audit sampling.", "Garnet data retention memo 173: The cache lifetime is capped at 9 days during Q1, before the regional handoff is signed. The note also says archived exceptions remain active.", "Garnet data retention memo 173: The cache lifetime target is at least 7 days during Q1, before the regional handoff is signed. Teams may exceed it when capacity is low."], "id": "meaning-q-39cacdcaac2e39a2", "positive_doc_id": "meaning-doc-ec9a17811d548ef4", "positive_text": "Garnet data retention memo 173: The cache lifetime is capped at 7 days during Q1, before the regional handoff is signed. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Garnet's data retention cache lifetime has a 7 days cap in Q1 before the regional handoff is signed, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "12 hours; after the second failed retry; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-5eb50e42cd0c133e", "meaning-doc-72dc4e7016799977", "meaning-doc-6f2158190e4d2263"], "hard_negative_texts": ["Fjord cloud billing memo 174: The escalation limit is capped at 12 hours during Q2, when the customer tier is starter. Archived exceptions are counted as active for audit sampling.", "Fjord cloud billing memo 174: The escalation limit is capped at 10 hours during Q2, after the second failed retry. The note also says archived exceptions remain active.", "Fjord cloud billing memo 174: The escalation limit target is at least 12 hours during Q2, after the second failed retry. Teams may exceed it when capacity is low."], "id": "meaning-q-b11c8f0711054f37", "positive_doc_id": "meaning-doc-713f947807094cd2", "positive_text": "Fjord cloud billing memo 174: The escalation limit is capped at 12 hours during Q2, after the second failed retry. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Fjord's cloud billing escalation limit has a 12 hours cap in Q2 after the second failed retry, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "18 percent; only when the request is marked external; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-e41a5b1d438f27ff", "meaning-doc-5650bc3d6def7574", "meaning-doc-cff7f7ee4bdcde10"], "hard_negative_texts": ["Cedar warehouse operations memo 175: The cache lifetime is capped at 18 percent during Q1, after the first successful retry. Archived exceptions are counted as active for audit sampling.", "Cedar warehouse operations memo 175: The cache lifetime is capped at 21 percent during Q1, only when the request is marked external. The note also says archived exceptions remain active.", "Cedar warehouse operations memo 175: The cache lifetime target is at least 18 percent during Q1, only when the request is marked external. Teams may exceed it when capacity is low."], "id": "meaning-q-a31ed46d7bc29300", "positive_doc_id": "meaning-doc-a7ccd45ba0a0e2a4", "positive_text": "Cedar warehouse operations memo 175: The cache lifetime is capped at 18 percent during Q1, only when the request is marked external. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Cedar's warehouse operations cache lifetime has a 18 percent cap in Q1 only when the request is marked external, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "12 tickets; only when the request is marked external; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-17b3c609f8e5ea1d", "meaning-doc-9c96f954721881a9", "meaning-doc-03e4952a012afe8f"], "hard_negative_texts": ["Juniper security review memo 176: The error budget is capped at 12 tickets during Q1, only when the owner is the platform team. Archived exceptions are counted as active for audit sampling.", "Juniper security review memo 176: The error budget is capped at 18 tickets during Q1, only when the request is marked external. The note also says archived exceptions remain active.", "Juniper security review memo 176: The error budget target is at least 12 tickets during Q1, only when the request is marked external. Teams may exceed it when capacity is low."], "id": "meaning-q-2c85f127a7e0683a", "positive_doc_id": "meaning-doc-817edbabfff6f535", "positive_text": "Juniper security review memo 176: The error budget is capped at 12 tickets during Q1, only when the request is marked external. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Juniper's security review error budget has a 12 tickets cap in Q1 only when the request is marked external, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "7 tickets; when the customer tier is enterprise; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-09f671ea2d11b7bc", "meaning-doc-37368fcbeeb03f56", "meaning-doc-c13215d011c1bdff"], "hard_negative_texts": ["Juniper payments memo 177: The latency is capped at 7 tickets during Q2, only when the owner is the platform team. Archived exceptions are counted as active for audit sampling.", "Juniper payments memo 177: The latency is capped at 13 tickets during Q2, when the customer tier is enterprise. The note also says archived exceptions remain active.", "Juniper payments memo 177: The latency target is at least 7 tickets during Q2, when the customer tier is enterprise. Teams may exceed it when capacity is low."], "id": "meaning-q-7de6c067eb0cb9a7", "positive_doc_id": "meaning-doc-1e589d7f239c5aaf", "positive_text": "Juniper payments memo 177: The latency is capped at 7 tickets during Q2, when the customer tier is enterprise. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Juniper's payments latency has a 7 tickets cap in Q2 when the customer tier is enterprise, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "96 percent; unless the owner is the platform team; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-fca5f8c57f56506d", "meaning-doc-3ca12a7bb1ed1fa2", "meaning-doc-9dc62480fc3ab6af"], "hard_negative_texts": ["Harbor payments memo 178: The approval threshold is capped at 96 percent during Q2, after the first successful retry. Archived exceptions are counted as active for audit sampling.", "Harbor payments memo 178: The approval threshold is capped at 99 percent during Q2, unless the owner is the platform team. The note also says archived exceptions remain active.", "Harbor payments memo 178: The approval threshold target is at least 96 percent during Q2, unless the owner is the platform team. Teams may exceed it when capacity is low."], "id": "meaning-q-802174f8df3ea63c", "positive_doc_id": "meaning-doc-1a887aa13c736716", "positive_text": "Harbor payments memo 178: The approval threshold is capped at 96 percent during Q2, unless the owner is the platform team. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Harbor's payments approval threshold has a 96 percent cap in Q2 unless the owner is the platform team, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "12 days; before the regional handoff is signed; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-bed56cdff3f485de", "meaning-doc-ba96cfb0f961c74a", "meaning-doc-1c98ba91c219d7dc"], "hard_negative_texts": ["Basil warehouse operations memo 179: The payment hold is capped at 12 days during Q3, after the regional handoff is signed. Archived exceptions are counted as active for audit sampling.", "Basil warehouse operations memo 179: The payment hold is capped at 10 days during Q3, before the regional handoff is signed. The note also says archived exceptions remain active.", "Basil warehouse operations memo 179: The payment hold target is at least 12 days during Q3, before the regional handoff is signed. Teams may exceed it when capacity is low."], "id": "meaning-q-dff5bf751b5ff195", "positive_doc_id": "meaning-doc-02d318b1b4c4c3da", "positive_text": "Basil warehouse operations memo 179: The payment hold is capped at 12 days during Q3, before the regional handoff is signed. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Basil's warehouse operations payment hold has a 12 days cap in Q3 before the regional handoff is signed, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "72 percent; before the regional handoff is signed; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-d57f1472a149470d", "meaning-doc-dac2bcde77a444d7", "meaning-doc-71751889b507feb9"], "hard_negative_texts": ["Indigo privacy requests memo 180: The cache lifetime is capped at 72 percent during Q3, for every request regardless of source. Archived exceptions are counted as active for audit sampling.", "Indigo privacy requests memo 180: The cache lifetime is capped at 70 percent during Q3, before the regional handoff is signed. The note also says archived exceptions remain active.", "Indigo privacy requests memo 180: The cache lifetime target is at least 72 percent during Q3, before the regional handoff is signed. Teams may exceed it when capacity is low."], "id": "meaning-q-7eb2247921fb8021", "positive_doc_id": "meaning-doc-c13472e07e4a6ad1", "positive_text": "Indigo privacy requests memo 180: The cache lifetime is capped at 72 percent during Q3, before the regional handoff is signed. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Indigo's privacy requests cache lifetime has a 72 percent cap in Q3 before the regional handoff is signed, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "96 minutes; before the regional handoff is signed; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-dcdce798154b05fc", "meaning-doc-b7854b70adbeb112", "meaning-doc-255dccb69d1f5924"], "hard_negative_texts": ["Fjord cloud billing memo 181: The payment hold is capped at 96 minutes during Q3, when the customer tier is starter. Archived exceptions are counted as active for audit sampling.", "Fjord cloud billing memo 181: The payment hold is capped at 93 minutes during Q3, before the regional handoff is signed. The note also says archived exceptions remain active.", "Fjord cloud billing memo 181: The payment hold target is at least 96 minutes during Q3, before the regional handoff is signed. Teams may exceed it when capacity is low."], "id": "meaning-q-731dfb2b8ee30ab0", "positive_doc_id": "meaning-doc-2bffa99f14256a7d", "positive_text": "Fjord cloud billing memo 181: The payment hold is capped at 96 minutes during Q3, before the regional handoff is signed. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Fjord's cloud billing payment hold has a 96 minutes cap in Q3 before the regional handoff is signed, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "18 requests; before the regional handoff is signed; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-954bca17ceaf5977", "meaning-doc-966b69dd6f4d12b3", "meaning-doc-55da316ed928721b"], "hard_negative_texts": ["Kepler model deployment memo 182: The escalation limit is capped at 18 requests during Q4, after the regional handoff is signed. Archived exceptions are counted as active for audit sampling.", "Kepler model deployment memo 182: The escalation limit is capped at 21 requests during Q4, before the regional handoff is signed. The note also says archived exceptions remain active.", "Kepler model deployment memo 182: The escalation limit target is at least 18 requests during Q4, before the regional handoff is signed. Teams may exceed it when capacity is low."], "id": "meaning-q-86d0a4f38e4a7653", "positive_doc_id": "meaning-doc-595e5dea57730a13", "positive_text": "Kepler model deployment memo 182: The escalation limit is capped at 18 requests during Q4, before the regional handoff is signed. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Kepler's model deployment escalation limit has a 18 requests cap in Q4 before the regional handoff is signed, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "48 percent; only when the request is marked external; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-a1e281ae2e41ef41", "meaning-doc-5534ce96d471dad5", "meaning-doc-44893f8a9648cfa6"], "hard_negative_texts": ["Fjord procurement memo 183: The payment hold is capped at 48 percent during Q1, after the first successful retry. Archived exceptions are counted as active for audit sampling.", "Fjord procurement memo 183: The payment hold is capped at 54 percent during Q1, only when the request is marked external. The note also says archived exceptions remain active.", "Fjord procurement memo 183: The payment hold target is at least 48 percent during Q1, only when the request is marked external. Teams may exceed it when capacity is low."], "id": "meaning-q-4e251e7923be1696", "positive_doc_id": "meaning-doc-9349cbc15ade2fbc", "positive_text": "Fjord procurement memo 183: The payment hold is capped at 48 percent during Q1, only when the request is marked external. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Fjord's procurement payment hold has a 48 percent cap in Q1 only when the request is marked external, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "72 requests; only when the request is marked external; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-b782e4f1852f8d96", "meaning-doc-048c41c98f232b23", "meaning-doc-58ba9adb1b8743a2"], "hard_negative_texts": ["Aurora model deployment memo 184: The approval threshold is capped at 72 requests during Q2, when the customer tier is starter. Archived exceptions are counted as active for audit sampling.", "Aurora model deployment memo 184: The approval threshold is capped at 74 requests during Q2, only when the request is marked external. The note also says archived exceptions remain active.", "Aurora model deployment memo 184: The approval threshold target is at least 72 requests during Q2, only when the request is marked external. Teams may exceed it when capacity is low."], "id": "meaning-q-13179be2b1fea4c9", "positive_doc_id": "meaning-doc-d9607ae645dccf14", "positive_text": "Aurora model deployment memo 184: The approval threshold is capped at 72 requests during Q2, only when the request is marked external. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Aurora's model deployment approval threshold has a 72 requests cap in Q2 only when the request is marked external, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "7 minutes; after the second failed retry; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-791fbe8b42f12f2f", "meaning-doc-5e97be71b63f7b57", "meaning-doc-ff509a70c908a2ce"], "hard_negative_texts": ["Kepler data retention memo 185: The cache lifetime is capped at 7 minutes during Q2, after the first successful retry. Archived exceptions are counted as active for audit sampling.", "Kepler data retention memo 185: The cache lifetime is capped at 10 minutes during Q2, after the second failed retry. The note also says archived exceptions remain active.", "Kepler data retention memo 185: The cache lifetime target is at least 7 minutes during Q2, after the second failed retry. Teams may exceed it when capacity is low."], "id": "meaning-q-35d3e9e5e09de08a", "positive_doc_id": "meaning-doc-e49eab8828eb4ab0", "positive_text": "Kepler data retention memo 185: The cache lifetime is capped at 7 minutes during Q2, after the second failed retry. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Kepler's data retention cache lifetime has a 7 minutes cap in Q2 after the second failed retry, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "36 hours; unless the owner is the platform team; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-5f87574fd1c54b27", "meaning-doc-4539df1ee12ede19", "meaning-doc-9e47bf88bcf1de9a"], "hard_negative_texts": ["Fjord cloud billing memo 186: The error budget is capped at 36 hours during Q2, when the customer tier is starter. Archived exceptions are counted as active for audit sampling.", "Fjord cloud billing memo 186: The error budget is capped at 38 hours during Q2, unless the owner is the platform team. The note also says archived exceptions remain active.", "Fjord cloud billing memo 186: The error budget target is at least 36 hours during Q2, unless the owner is the platform team. Teams may exceed it when capacity is low."], "id": "meaning-q-f6f34ca1d72956ad", "positive_doc_id": "meaning-doc-7a11f8731f12782b", "positive_text": "Fjord cloud billing memo 186: The error budget is capped at 36 hours during Q2, unless the owner is the platform team. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Fjord's cloud billing error budget has a 36 hours cap in Q2 unless the owner is the platform team, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "96 hours; after the second failed retry; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-ac672e8e3602a5f8", "meaning-doc-3116a0ce33317160", "meaning-doc-dd053a1e569c6f32"], "hard_negative_texts": ["Indigo clinical triage memo 187: The escalation limit is capped at 96 hours during Q3, after the first successful retry. Archived exceptions are counted as active for audit sampling.", "Indigo clinical triage memo 187: The escalation limit is capped at 94 hours during Q3, after the second failed retry. The note also says archived exceptions remain active.", "Indigo clinical triage memo 187: The escalation limit target is at least 96 hours during Q3, after the second failed retry. Teams may exceed it when capacity is low."], "id": "meaning-q-216583931a543643", "positive_doc_id": "meaning-doc-7186f9eebe509d47", "positive_text": "Indigo clinical triage memo 187: The escalation limit is capped at 96 hours during Q3, after the second failed retry. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Indigo's clinical triage escalation limit has a 96 hours cap in Q3 after the second failed retry, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "7 tickets; after the second failed retry; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-9242e3ddc925a9b3", "meaning-doc-9b1d33b40c5901aa", "meaning-doc-b2b3b6e549656039"], "hard_negative_texts": ["Kepler release management memo 188: The cache lifetime is capped at 7 tickets during Q4, when the customer tier is starter. Archived exceptions are counted as active for audit sampling.", "Kepler release management memo 188: The cache lifetime is capped at 5 tickets during Q4, after the second failed retry. The note also says archived exceptions remain active.", "Kepler release management memo 188: The cache lifetime target is at least 7 tickets during Q4, after the second failed retry. Teams may exceed it when capacity is low."], "id": "meaning-q-ed5a535aa2f5abf7", "positive_doc_id": "meaning-doc-3737e2670d73f152", "positive_text": "Kepler release management memo 188: The cache lifetime is capped at 7 tickets during Q4, after the second failed retry. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Kepler's release management cache lifetime has a 7 tickets cap in Q4 after the second failed retry, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "24 requests; after the second failed retry; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-498319a0fccd6ca4", "meaning-doc-669348eae517d097", "meaning-doc-77fc2c04dd8bc478"], "hard_negative_texts": ["Ember edge caching memo 189: The error budget is capped at 24 requests during Q3, only when the owner is the platform team. Archived exceptions are counted as active for audit sampling.", "Ember edge caching memo 189: The error budget is capped at 27 requests during Q3, after the second failed retry. The note also says archived exceptions remain active.", "Ember edge caching memo 189: The error budget target is at least 24 requests during Q3, after the second failed retry. Teams may exceed it when capacity is low."], "id": "meaning-q-e63c3f4a81ec27b9", "positive_doc_id": "meaning-doc-de25bf99b7b8bf62", "positive_text": "Ember edge caching memo 189: The error budget is capped at 24 requests during Q3, after the second failed retry. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Ember's edge caching error budget has a 24 requests cap in Q3 after the second failed retry, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "96 hours; before the regional handoff is signed; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-1ab0afe82c0ccf5f", "meaning-doc-8cf6cd6a85c78bd2", "meaning-doc-d32289357e8a9383"], "hard_negative_texts": ["Delta privacy requests memo 190: The review level is capped at 96 hours during Q4, after the regional handoff is signed. Archived exceptions are counted as active for audit sampling.", "Delta privacy requests memo 190: The review level is capped at 102 hours during Q4, before the regional handoff is signed. The note also says archived exceptions remain active.", "Delta privacy requests memo 190: The review level target is at least 96 hours during Q4, before the regional handoff is signed. Teams may exceed it when capacity is low."], "id": "meaning-q-1fce4b47cf3f1491", "positive_doc_id": "meaning-doc-ef915ddf8079f1ab", "positive_text": "Delta privacy requests memo 190: The review level is capped at 96 hours during Q4, before the regional handoff is signed. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Delta's privacy requests review level has a 96 hours cap in Q4 before the regional handoff is signed, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "96 minutes; unless the owner is the platform team; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-ab01e29169d3ef9c", "meaning-doc-b21c6b596e843c21", "meaning-doc-0a58bdd7b8467150"], "hard_negative_texts": ["Ember warehouse operations memo 191: The sampling rate is capped at 96 minutes during Q3, only when the owner is the platform team. Archived exceptions are counted as active for audit sampling.", "Ember warehouse operations memo 191: The sampling rate is capped at 99 minutes during Q3, unless the owner is the platform team. The note also says archived exceptions remain active.", "Ember warehouse operations memo 191: The sampling rate target is at least 96 minutes during Q3, unless the owner is the platform team. Teams may exceed it when capacity is low."], "id": "meaning-q-5c43ea9d3bd97550", "positive_doc_id": "meaning-doc-1d323ecd565b5c7a", "positive_text": "Ember warehouse operations memo 191: The sampling rate is capped at 96 minutes during Q3, unless the owner is the platform team. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Ember's warehouse operations sampling rate has a 96 minutes cap in Q3 unless the owner is the platform team, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "12 minutes; only when the request is marked external; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-936f7d1ca0497224", "meaning-doc-6365f50546c841cb", "meaning-doc-df0709eac18cdb24"], "hard_negative_texts": ["Cedar privacy requests memo 192: The escalation limit is capped at 12 minutes during Q3, only when the owner is the platform team. Archived exceptions are counted as active for audit sampling.", "Cedar privacy requests memo 192: The escalation limit is capped at 15 minutes during Q3, only when the request is marked external. The note also says archived exceptions remain active.", "Cedar privacy requests memo 192: The escalation limit target is at least 12 minutes during Q3, only when the request is marked external. Teams may exceed it when capacity is low."], "id": "meaning-q-74621da56550385c", "positive_doc_id": "meaning-doc-5ad95c2a575a661f", "positive_text": "Cedar privacy requests memo 192: The escalation limit is capped at 12 minutes during Q3, only when the request is marked external. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Cedar's privacy requests escalation limit has a 12 minutes cap in Q3 only when the request is marked external, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "36 hours; unless the owner is the platform team; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-3112462f7fa6509e", "meaning-doc-a0682670f9bd0a2b", "meaning-doc-8eb0302a1a55058e"], "hard_negative_texts": ["Garnet warehouse operations memo 193: The approval threshold is capped at 36 hours during Q2, after the regional handoff is signed. Archived exceptions are counted as active for audit sampling.", "Garnet warehouse operations memo 193: The approval threshold is capped at 33 hours during Q2, unless the owner is the platform team. The note also says archived exceptions remain active.", "Garnet warehouse operations memo 193: The approval threshold target is at least 36 hours during Q2, unless the owner is the platform team. Teams may exceed it when capacity is low."], "id": "meaning-q-b1e6e25e6733e668", "positive_doc_id": "meaning-doc-541411b5f2299afd", "positive_text": "Garnet warehouse operations memo 193: The approval threshold is capped at 36 hours during Q2, unless the owner is the platform team. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Garnet's warehouse operations approval threshold has a 36 hours cap in Q2 unless the owner is the platform team, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "96 hours; unless the owner is the platform team; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-2c8338b27af72e64", "meaning-doc-30c2d179f59bbff1", "meaning-doc-d837da43a8d2bd95"], "hard_negative_texts": ["Ember privacy requests memo 194: The cache lifetime is capped at 96 hours during Q4, only when the owner is the platform team. Archived exceptions are counted as active for audit sampling.", "Ember privacy requests memo 194: The cache lifetime is capped at 94 hours during Q4, unless the owner is the platform team. The note also says archived exceptions remain active.", "Ember privacy requests memo 194: The cache lifetime target is at least 96 hours during Q4, unless the owner is the platform team. Teams may exceed it when capacity is low."], "id": "meaning-q-efa84a028b2ec762", "positive_doc_id": "meaning-doc-e5908b8218f49b09", "positive_text": "Ember privacy requests memo 194: The cache lifetime is capped at 96 hours during Q4, unless the owner is the platform team. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Ember's privacy requests cache lifetime has a 96 hours cap in Q4 unless the owner is the platform team, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "48 tickets; after the second failed retry; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-ec5f0aa9c0ed7382", "meaning-doc-d49428eaba69193f", "meaning-doc-03c60cb46a4500c4"], "hard_negative_texts": ["Kepler payments memo 195: The error budget is capped at 48 tickets during Q1, after the first successful retry. Archived exceptions are counted as active for audit sampling.", "Kepler payments memo 195: The error budget is capped at 50 tickets during Q1, after the second failed retry. The note also says archived exceptions remain active.", "Kepler payments memo 195: The error budget target is at least 48 tickets during Q1, after the second failed retry. Teams may exceed it when capacity is low."], "id": "meaning-q-cfaaf28e70dcb2fe", "positive_doc_id": "meaning-doc-bd446ad953bfb7a2", "positive_text": "Kepler payments memo 195: The error budget is capped at 48 tickets during Q1, after the second failed retry. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Kepler's payments error budget has a 48 tickets cap in Q1 after the second failed retry, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "48 tickets; unless the owner is the platform team; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-b70fbe93ceeeb5ea", "meaning-doc-1598a0308274eaf0", "meaning-doc-55642f4587d1a0b5"], "hard_negative_texts": ["Delta warehouse operations memo 196: The approval threshold is capped at 48 tickets during Q3, after the first successful retry. Archived exceptions are counted as active for audit sampling.", "Delta warehouse operations memo 196: The approval threshold is capped at 45 tickets during Q3, unless the owner is the platform team. The note also says archived exceptions remain active.", "Delta warehouse operations memo 196: The approval threshold target is at least 48 tickets during Q3, unless the owner is the platform team. Teams may exceed it when capacity is low."], "id": "meaning-q-12c2832d9a478d73", "positive_doc_id": "meaning-doc-153af5b0257ff81c", "positive_text": "Delta warehouse operations memo 196: The approval threshold is capped at 48 tickets during Q3, unless the owner is the platform team. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Delta's warehouse operations approval threshold has a 48 tickets cap in Q3 unless the owner is the platform team, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "12 requests; when the customer tier is enterprise; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-62f94630f0b50d54", "meaning-doc-b831de28bbfeac57", "meaning-doc-7ea599e0645d41fb"], "hard_negative_texts": ["Kepler incident response memo 197: The retention period is capped at 12 requests during Q3, after the regional handoff is signed. Archived exceptions are counted as active for audit sampling.", "Kepler incident response memo 197: The retention period is capped at 18 requests during Q3, when the customer tier is enterprise. The note also says archived exceptions remain active.", "Kepler incident response memo 197: The retention period target is at least 12 requests during Q3, when the customer tier is enterprise. Teams may exceed it when capacity is low."], "id": "meaning-q-862423416fd11a7d", "positive_doc_id": "meaning-doc-aeb65983bf9b37b4", "positive_text": "Kepler incident response memo 197: The retention period is capped at 12 requests during Q3, when the customer tier is enterprise. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Kepler's incident response retention period has a 12 requests cap in Q3 when the customer tier is enterprise, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "12 days; after the second failed retry; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-c821ed08244b5032", "meaning-doc-c38e2034ece9a91d", "meaning-doc-141abb84bcacf357"], "hard_negative_texts": ["Delta data retention memo 198: The sampling rate is capped at 12 days during Q4, after the regional handoff is signed. Archived exceptions are counted as active for audit sampling.", "Delta data retention memo 198: The sampling rate is capped at 10 days during Q4, after the second failed retry. The note also says archived exceptions remain active.", "Delta data retention memo 198: The sampling rate target is at least 12 days during Q4, after the second failed retry. Teams may exceed it when capacity is low."], "id": "meaning-q-8691d7e2e7a3ec26", "positive_doc_id": "meaning-doc-78e768efeb1dcc07", "positive_text": "Delta data retention memo 198: The sampling rate is capped at 12 days during Q4, after the second failed retry. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Delta's data retention sampling rate has a 12 days cap in Q4 after the second failed retry, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "96 minutes; unless the owner is the platform team; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-c4ad62fb54f95206", "meaning-doc-247450b87ed4471b", "meaning-doc-3041b2344f3648d7"], "hard_negative_texts": ["Garnet model deployment memo 199: The sampling rate is capped at 96 minutes during Q4, for every request regardless of source. Archived exceptions are counted as active for audit sampling.", "Garnet model deployment memo 199: The sampling rate is capped at 99 minutes during Q4, unless the owner is the platform team. The note also says archived exceptions remain active.", "Garnet model deployment memo 199: The sampling rate target is at least 96 minutes during Q4, unless the owner is the platform team. Teams may exceed it when capacity is low."], "id": "meaning-q-366edbfc3623b9ac", "positive_doc_id": "meaning-doc-1300bbce0707aa76", "positive_text": "Garnet model deployment memo 199: The sampling rate is capped at 96 minutes during Q4, unless the owner is the platform team. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Garnet's model deployment sampling rate has a 96 minutes cap in Q4 unless the owner is the platform team, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "72 minutes; when the customer tier is enterprise; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-a71df9c6993585cd", "meaning-doc-6ceff27e68ec7d27", "meaning-doc-7e16d3c20ab14439"], "hard_negative_texts": ["Basil data retention memo 200: The payment hold is capped at 72 minutes during Q4, after the regional handoff is signed. Archived exceptions are counted as active for audit sampling.", "Basil data retention memo 200: The payment hold is capped at 69 minutes during Q4, when the customer tier is enterprise. The note also says archived exceptions remain active.", "Basil data retention memo 200: The payment hold target is at least 72 minutes during Q4, when the customer tier is enterprise. Teams may exceed it when capacity is low."], "id": "meaning-q-204b543ca0feda57", "positive_doc_id": "meaning-doc-cba2f73ddf2f116d", "positive_text": "Basil data retention memo 200: The payment hold is capped at 72 minutes during Q4, when the customer tier is enterprise. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Basil's data retention payment hold has a 72 minutes cap in Q4 when the customer tier is enterprise, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "24 requests; when the customer tier is enterprise; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-20299db42f59f7f4", "meaning-doc-2630e7038021d66c", "meaning-doc-6aac74ff33d935a2"], "hard_negative_texts": ["Basil privacy requests memo 201: The latency is capped at 24 requests during Q3, for every request regardless of source. Archived exceptions are counted as active for audit sampling.", "Basil privacy requests memo 201: The latency is capped at 30 requests during Q3, when the customer tier is enterprise. The note also says archived exceptions remain active.", "Basil privacy requests memo 201: The latency target is at least 24 requests during Q3, when the customer tier is enterprise. Teams may exceed it when capacity is low."], "id": "meaning-q-f0c6c229105e2892", "positive_doc_id": "meaning-doc-4ebe03ab0482c2b8", "positive_text": "Basil privacy requests memo 201: The latency is capped at 24 requests during Q3, when the customer tier is enterprise. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Basil's privacy requests latency has a 24 requests cap in Q3 when the customer tier is enterprise, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "7 requests; unless the owner is the platform team; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-68c03dda646dec98", "meaning-doc-7ef0f06443ac2569", "meaning-doc-0d8ef9d3b880e19e"], "hard_negative_texts": ["Cedar privacy requests memo 202: The latency is capped at 7 requests during Q3, when the customer tier is starter. Archived exceptions are counted as active for audit sampling.", "Cedar privacy requests memo 202: The latency is capped at 10 requests during Q3, unless the owner is the platform team. The note also says archived exceptions remain active.", "Cedar privacy requests memo 202: The latency target is at least 7 requests during Q3, unless the owner is the platform team. Teams may exceed it when capacity is low."], "id": "meaning-q-5d769b054e41f45f", "positive_doc_id": "meaning-doc-dd9745f2c9826351", "positive_text": "Cedar privacy requests memo 202: The latency is capped at 7 requests during Q3, unless the owner is the platform team. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Cedar's privacy requests latency has a 7 requests cap in Q3 unless the owner is the platform team, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "7 hours; unless the owner is the platform team; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-8d419371a35b9438", "meaning-doc-0f1a5da7d6b3d3f6", "meaning-doc-b741a8f0b4eb7415"], "hard_negative_texts": ["Cedar edge caching memo 203: The cache lifetime is capped at 7 hours during Q3, after the regional handoff is signed. Archived exceptions are counted as active for audit sampling.", "Cedar edge caching memo 203: The cache lifetime is capped at 10 hours during Q3, unless the owner is the platform team. The note also says archived exceptions remain active.", "Cedar edge caching memo 203: The cache lifetime target is at least 7 hours during Q3, unless the owner is the platform team. Teams may exceed it when capacity is low."], "id": "meaning-q-5d8fa0a11dc0d5d0", "positive_doc_id": "meaning-doc-a10d68244f945a24", "positive_text": "Cedar edge caching memo 203: The cache lifetime is capped at 7 hours during Q3, unless the owner is the platform team. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Cedar's edge caching cache lifetime has a 7 hours cap in Q3 unless the owner is the platform team, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "7 tickets; when the customer tier is enterprise; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-aceb6fe3d5545292", "meaning-doc-c0ca56c3be742bda", "meaning-doc-adff8be11117a253"], "hard_negative_texts": ["Lumen procurement memo 204: The sampling rate is capped at 7 tickets during Q1, for every request regardless of source. Archived exceptions are counted as active for audit sampling.", "Lumen procurement memo 204: The sampling rate is capped at 5 tickets during Q1, when the customer tier is enterprise. The note also says archived exceptions remain active.", "Lumen procurement memo 204: The sampling rate target is at least 7 tickets during Q1, when the customer tier is enterprise. Teams may exceed it when capacity is low."], "id": "meaning-q-bc1e7e83663c3e1f", "positive_doc_id": "meaning-doc-59896858eef4f59a", "positive_text": "Lumen procurement memo 204: The sampling rate is capped at 7 tickets during Q1, when the customer tier is enterprise. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Lumen's procurement sampling rate has a 7 tickets cap in Q1 when the customer tier is enterprise, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "18 tickets; when the customer tier is enterprise; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-8d52e4cc3ffa7691", "meaning-doc-454381eff3e07cdd", "meaning-doc-82947d1defed63bc"], "hard_negative_texts": ["Aurora privacy requests memo 205: The cache lifetime is capped at 18 tickets during Q4, when the customer tier is starter. Archived exceptions are counted as active for audit sampling.", "Aurora privacy requests memo 205: The cache lifetime is capped at 16 tickets during Q4, when the customer tier is enterprise. The note also says archived exceptions remain active.", "Aurora privacy requests memo 205: The cache lifetime target is at least 18 tickets during Q4, when the customer tier is enterprise. Teams may exceed it when capacity is low."], "id": "meaning-q-7fe0629b3c967ed8", "positive_doc_id": "meaning-doc-8506c3b93661322d", "positive_text": "Aurora privacy requests memo 205: The cache lifetime is capped at 18 tickets during Q4, when the customer tier is enterprise. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Aurora's privacy requests cache lifetime has a 18 tickets cap in Q4 when the customer tier is enterprise, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "12 minutes; when the customer tier is enterprise; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-fc69d835253f341b", "meaning-doc-93d2fd210c2895d0", "meaning-doc-6bd39e68d8ddbfaa"], "hard_negative_texts": ["Harbor procurement memo 206: The latency is capped at 12 minutes during Q3, after the first successful retry. Archived exceptions are counted as active for audit sampling.", "Harbor procurement memo 206: The latency is capped at 10 minutes during Q3, when the customer tier is enterprise. The note also says archived exceptions remain active.", "Harbor procurement memo 206: The latency target is at least 12 minutes during Q3, when the customer tier is enterprise. Teams may exceed it when capacity is low."], "id": "meaning-q-dd41193a531342aa", "positive_doc_id": "meaning-doc-2207be17af051ede", "positive_text": "Harbor procurement memo 206: The latency is capped at 12 minutes during Q3, when the customer tier is enterprise. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Harbor's procurement latency has a 12 minutes cap in Q3 when the customer tier is enterprise, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "96 minutes; when the customer tier is enterprise; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-a9fe9b1e58d6b6e9", "meaning-doc-dd6bf37f554d2cce", "meaning-doc-9427151e0dcd3ce7"], "hard_negative_texts": ["Kepler payments memo 207: The payment hold is capped at 96 minutes during Q3, for every request regardless of source. Archived exceptions are counted as active for audit sampling.", "Kepler payments memo 207: The payment hold is capped at 99 minutes during Q3, when the customer tier is enterprise. The note also says archived exceptions remain active.", "Kepler payments memo 207: The payment hold target is at least 96 minutes during Q3, when the customer tier is enterprise. Teams may exceed it when capacity is low."], "id": "meaning-q-a16043d177b72418", "positive_doc_id": "meaning-doc-06b8ed5dea466410", "positive_text": "Kepler payments memo 207: The payment hold is capped at 96 minutes during Q3, when the customer tier is enterprise. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Kepler's payments payment hold has a 96 minutes cap in Q3 when the customer tier is enterprise, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "36 minutes; unless the owner is the platform team; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-6ecf0126e589e317", "meaning-doc-6d399c5e97bee4a9", "meaning-doc-6fd21cc33ea930df"], "hard_negative_texts": ["Indigo payments memo 208: The escalation limit is capped at 36 minutes during Q3, when the customer tier is starter. Archived exceptions are counted as active for audit sampling.", "Indigo payments memo 208: The escalation limit is capped at 42 minutes during Q3, unless the owner is the platform team. The note also says archived exceptions remain active.", "Indigo payments memo 208: The escalation limit target is at least 36 minutes during Q3, unless the owner is the platform team. Teams may exceed it when capacity is low."], "id": "meaning-q-45183b64671219ec", "positive_doc_id": "meaning-doc-5666c6ab404fc0f3", "positive_text": "Indigo payments memo 208: The escalation limit is capped at 36 minutes during Q3, unless the owner is the platform team. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Indigo's payments escalation limit has a 36 minutes cap in Q3 unless the owner is the platform team, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "72 percent; unless the owner is the platform team; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-fae4ddebdcbf26cb", "meaning-doc-a769c741477e9742", "meaning-doc-d9a2c3434539348f"], "hard_negative_texts": ["Cedar security review memo 209: The review level is capped at 72 percent during Q2, only when the owner is the platform team. Archived exceptions are counted as active for audit sampling.", "Cedar security review memo 209: The review level is capped at 69 percent during Q2, unless the owner is the platform team. The note also says archived exceptions remain active.", "Cedar security review memo 209: The review level target is at least 72 percent during Q2, unless the owner is the platform team. Teams may exceed it when capacity is low."], "id": "meaning-q-7e6e9e9338fb7250", "positive_doc_id": "meaning-doc-3a27aa0513dc965d", "positive_text": "Cedar security review memo 209: The review level is capped at 72 percent during Q2, unless the owner is the platform team. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Cedar's security review review level has a 72 percent cap in Q2 unless the owner is the platform team, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "96 hours; before the regional handoff is signed; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-2a0459594c2f29f4", "meaning-doc-0f56b22d9cedb694", "meaning-doc-f48f3b3357baf00f"], "hard_negative_texts": ["Aurora release management memo 210: The approval threshold is capped at 96 hours during Q3, when the customer tier is starter. Archived exceptions are counted as active for audit sampling.", "Aurora release management memo 210: The approval threshold is capped at 99 hours during Q3, before the regional handoff is signed. The note also says archived exceptions remain active.", "Aurora release management memo 210: The approval threshold target is at least 96 hours during Q3, before the regional handoff is signed. Teams may exceed it when capacity is low."], "id": "meaning-q-134c3d4655d051da", "positive_doc_id": "meaning-doc-755a3227fb65b7e1", "positive_text": "Aurora release management memo 210: The approval threshold is capped at 96 hours during Q3, before the regional handoff is signed. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Aurora's release management approval threshold has a 96 hours cap in Q3 before the regional handoff is signed, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "7 percent; only when the request is marked external; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-fa0facb3997fa67b", "meaning-doc-c7ad5d7a2f00e087", "meaning-doc-098e57dc078b3714"], "hard_negative_texts": ["Aurora clinical triage memo 211: The escalation limit is capped at 7 percent during Q3, after the regional handoff is signed. Archived exceptions are counted as active for audit sampling.", "Aurora clinical triage memo 211: The escalation limit is capped at 5 percent during Q3, only when the request is marked external. The note also says archived exceptions remain active.", "Aurora clinical triage memo 211: The escalation limit target is at least 7 percent during Q3, only when the request is marked external. Teams may exceed it when capacity is low."], "id": "meaning-q-4843b2f30a32bb09", "positive_doc_id": "meaning-doc-525544dc15c886b5", "positive_text": "Aurora clinical triage memo 211: The escalation limit is capped at 7 percent during Q3, only when the request is marked external. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Aurora's clinical triage escalation limit has a 7 percent cap in Q3 only when the request is marked external, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "7 requests; after the second failed retry; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-d69f8eaee34a6574", "meaning-doc-467e78b57ff7a016", "meaning-doc-bd300465ef9d5819"], "hard_negative_texts": ["Basil edge caching memo 212: The approval threshold is capped at 7 requests during Q1, only when the owner is the platform team. Archived exceptions are counted as active for audit sampling.", "Basil edge caching memo 212: The approval threshold is capped at 5 requests during Q1, after the second failed retry. The note also says archived exceptions remain active.", "Basil edge caching memo 212: The approval threshold target is at least 7 requests during Q1, after the second failed retry. Teams may exceed it when capacity is low."], "id": "meaning-q-ca54dec4a784c248", "positive_doc_id": "meaning-doc-b611ab1103dba058", "positive_text": "Basil edge caching memo 212: The approval threshold is capped at 7 requests during Q1, after the second failed retry. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Basil's edge caching approval threshold has a 7 requests cap in Q1 after the second failed retry, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "36 requests; only when the request is marked external; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-8eb8004dbf16f5bf", "meaning-doc-a73c9c438517ba56", "meaning-doc-7c2a115eaac1a5de"], "hard_negative_texts": ["Ember clinical triage memo 213: The approval threshold is capped at 36 requests during Q2, for every request regardless of source. Archived exceptions are counted as active for audit sampling.", "Ember clinical triage memo 213: The approval threshold is capped at 34 requests during Q2, only when the request is marked external. The note also says archived exceptions remain active.", "Ember clinical triage memo 213: The approval threshold target is at least 36 requests during Q2, only when the request is marked external. Teams may exceed it when capacity is low."], "id": "meaning-q-bea03c8207bfcedd", "positive_doc_id": "meaning-doc-ca68c1553e2f7ce6", "positive_text": "Ember clinical triage memo 213: The approval threshold is capped at 36 requests during Q2, only when the request is marked external. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Ember's clinical triage approval threshold has a 36 requests cap in Q2 only when the request is marked external, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "36 requests; after the second failed retry; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-5f666fd568f3c3f0", "meaning-doc-b059fdeac721d526", "meaning-doc-bb51bb233433aeff"], "hard_negative_texts": ["Fjord model deployment memo 214: The cache lifetime is capped at 36 requests during Q4, after the regional handoff is signed. Archived exceptions are counted as active for audit sampling.", "Fjord model deployment memo 214: The cache lifetime is capped at 38 requests during Q4, after the second failed retry. The note also says archived exceptions remain active.", "Fjord model deployment memo 214: The cache lifetime target is at least 36 requests during Q4, after the second failed retry. Teams may exceed it when capacity is low."], "id": "meaning-q-b3ed478cdd60c174", "positive_doc_id": "meaning-doc-afb67295d61d47e2", "positive_text": "Fjord model deployment memo 214: The cache lifetime is capped at 36 requests during Q4, after the second failed retry. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Fjord's model deployment cache lifetime has a 36 requests cap in Q4 after the second failed retry, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "24 tickets; when the customer tier is enterprise; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-f754599d872cfc0d", "meaning-doc-eceebb49d605dc30", "meaning-doc-a49ba8bf7a92fb91"], "hard_negative_texts": ["Indigo data retention memo 215: The cache lifetime is capped at 24 tickets during Q4, after the first successful retry. Archived exceptions are counted as active for audit sampling.", "Indigo data retention memo 215: The cache lifetime is capped at 26 tickets during Q4, when the customer tier is enterprise. The note also says archived exceptions remain active.", "Indigo data retention memo 215: The cache lifetime target is at least 24 tickets during Q4, when the customer tier is enterprise. Teams may exceed it when capacity is low."], "id": "meaning-q-f35c89a712a1a7e1", "positive_doc_id": "meaning-doc-1dad023e222d84d1", "positive_text": "Indigo data retention memo 215: The cache lifetime is capped at 24 tickets during Q4, when the customer tier is enterprise. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Indigo's data retention cache lifetime has a 24 tickets cap in Q4 when the customer tier is enterprise, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "7 requests; when the customer tier is enterprise; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-70996c5cfd432e26", "meaning-doc-13781c3e1166a31c", "meaning-doc-392d010a9d65d649"], "hard_negative_texts": ["Harbor cloud billing memo 216: The escalation limit is capped at 7 requests during Q3, after the first successful retry. Archived exceptions are counted as active for audit sampling.", "Harbor cloud billing memo 216: The escalation limit is capped at 5 requests during Q3, when the customer tier is enterprise. The note also says archived exceptions remain active.", "Harbor cloud billing memo 216: The escalation limit target is at least 7 requests during Q3, when the customer tier is enterprise. Teams may exceed it when capacity is low."], "id": "meaning-q-43863f8902eb6588", "positive_doc_id": "meaning-doc-8d11484f7423b979", "positive_text": "Harbor cloud billing memo 216: The escalation limit is capped at 7 requests during Q3, when the customer tier is enterprise. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Harbor's cloud billing escalation limit has a 7 requests cap in Q3 when the customer tier is enterprise, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "12 percent; only when the request is marked external; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-54ba317c44f75e00", "meaning-doc-68b38e9e867a7466", "meaning-doc-e61067a148c287ff"], "hard_negative_texts": ["Juniper procurement memo 217: The latency is capped at 12 percent during Q4, after the regional handoff is signed. Archived exceptions are counted as active for audit sampling.", "Juniper procurement memo 217: The latency is capped at 15 percent during Q4, only when the request is marked external. The note also says archived exceptions remain active.", "Juniper procurement memo 217: The latency target is at least 12 percent during Q4, only when the request is marked external. Teams may exceed it when capacity is low."], "id": "meaning-q-db62472ea1a899c6", "positive_doc_id": "meaning-doc-db11b94d7f40c557", "positive_text": "Juniper procurement memo 217: The latency is capped at 12 percent during Q4, only when the request is marked external. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Juniper's procurement latency has a 12 percent cap in Q4 only when the request is marked external, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "18 percent; when the customer tier is enterprise; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-885c726e2a3461fb", "meaning-doc-073d79429409d7f4", "meaning-doc-75f9401cd2269273"], "hard_negative_texts": ["Juniper cloud billing memo 218: The retention period is capped at 18 percent during Q4, only when the owner is the platform team. Archived exceptions are counted as active for audit sampling.", "Juniper cloud billing memo 218: The retention period is capped at 15 percent during Q4, when the customer tier is enterprise. The note also says archived exceptions remain active.", "Juniper cloud billing memo 218: The retention period target is at least 18 percent during Q4, when the customer tier is enterprise. Teams may exceed it when capacity is low."], "id": "meaning-q-579f4407d8f69148", "positive_doc_id": "meaning-doc-61175c99c164c86b", "positive_text": "Juniper cloud billing memo 218: The retention period is capped at 18 percent during Q4, when the customer tier is enterprise. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Juniper's cloud billing retention period has a 18 percent cap in Q4 when the customer tier is enterprise, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "24 percent; unless the owner is the platform team; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-54a2036433e062ab", "meaning-doc-a21b35f25a516d77", "meaning-doc-67501218a724603c"], "hard_negative_texts": ["Ember incident response memo 219: The payment hold is capped at 24 percent during Q2, when the customer tier is starter. Archived exceptions are counted as active for audit sampling.", "Ember incident response memo 219: The payment hold is capped at 27 percent during Q2, unless the owner is the platform team. The note also says archived exceptions remain active.", "Ember incident response memo 219: The payment hold target is at least 24 percent during Q2, unless the owner is the platform team. Teams may exceed it when capacity is low."], "id": "meaning-q-26ac8b284d258e86", "positive_doc_id": "meaning-doc-567af7fa48267202", "positive_text": "Ember incident response memo 219: The payment hold is capped at 24 percent during Q2, unless the owner is the platform team. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Ember's incident response payment hold has a 24 percent cap in Q2 unless the owner is the platform team, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "72 hours; only when the request is marked external; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-ffb8546778e629a7", "meaning-doc-e19efc72075c1691", "meaning-doc-75efef2bed70fe0f"], "hard_negative_texts": ["Fjord clinical triage memo 220: The review level is capped at 72 hours during Q3, only when the owner is the platform team. Archived exceptions are counted as active for audit sampling.", "Fjord clinical triage memo 220: The review level is capped at 69 hours during Q3, only when the request is marked external. The note also says archived exceptions remain active.", "Fjord clinical triage memo 220: The review level target is at least 72 hours during Q3, only when the request is marked external. Teams may exceed it when capacity is low."], "id": "meaning-q-43c95a611f702281", "positive_doc_id": "meaning-doc-94d719022ce65265", "positive_text": "Fjord clinical triage memo 220: The review level is capped at 72 hours during Q3, only when the request is marked external. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Fjord's clinical triage review level has a 72 hours cap in Q3 only when the request is marked external, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "36 hours; when the customer tier is enterprise; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-5ef2bbc1d7bcfdd9", "meaning-doc-95b3d207460080a8", "meaning-doc-60dc217cdf71a887"], "hard_negative_texts": ["Fjord incident response memo 221: The error budget is capped at 36 hours during Q2, only when the owner is the platform team. Archived exceptions are counted as active for audit sampling.", "Fjord incident response memo 221: The error budget is capped at 39 hours during Q2, when the customer tier is enterprise. The note also says archived exceptions remain active.", "Fjord incident response memo 221: The error budget target is at least 36 hours during Q2, when the customer tier is enterprise. Teams may exceed it when capacity is low."], "id": "meaning-q-a2ffbcc992d45f76", "positive_doc_id": "meaning-doc-278ad32865fc2ae9", "positive_text": "Fjord incident response memo 221: The error budget is capped at 36 hours during Q2, when the customer tier is enterprise. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Fjord's incident response error budget has a 36 hours cap in Q2 when the customer tier is enterprise, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "18 minutes; only when the request is marked external; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-09734aa1d9dc2891", "meaning-doc-a764068f0fcf8dab", "meaning-doc-17386e41516dc12d"], "hard_negative_texts": ["Indigo warehouse operations memo 222: The retention period is capped at 18 minutes during Q4, only when the owner is the platform team. Archived exceptions are counted as active for audit sampling.", "Indigo warehouse operations memo 222: The retention period is capped at 15 minutes during Q4, only when the request is marked external. The note also says archived exceptions remain active.", "Indigo warehouse operations memo 222: The retention period target is at least 18 minutes during Q4, only when the request is marked external. Teams may exceed it when capacity is low."], "id": "meaning-q-e7db24a81959a0db", "positive_doc_id": "meaning-doc-707532bc0b2124b3", "positive_text": "Indigo warehouse operations memo 222: The retention period is capped at 18 minutes during Q4, only when the request is marked external. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Indigo's warehouse operations retention period has a 18 minutes cap in Q4 only when the request is marked external, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "24 minutes; after the second failed retry; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-bcce6e954c9be047", "meaning-doc-10b031eec1ef5e53", "meaning-doc-1fa3cdd87eaa02da"], "hard_negative_texts": ["Delta privacy requests memo 223: The review level is capped at 24 minutes during Q3, for every request regardless of source. Archived exceptions are counted as active for audit sampling.", "Delta privacy requests memo 223: The review level is capped at 22 minutes during Q3, after the second failed retry. The note also says archived exceptions remain active.", "Delta privacy requests memo 223: The review level target is at least 24 minutes during Q3, after the second failed retry. Teams may exceed it when capacity is low."], "id": "meaning-q-f7d4ddbb7e9249bb", "positive_doc_id": "meaning-doc-a761e881589513f3", "positive_text": "Delta privacy requests memo 223: The review level is capped at 24 minutes during Q3, after the second failed retry. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Delta's privacy requests review level has a 24 minutes cap in Q3 after the second failed retry, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "48 days; when the customer tier is enterprise; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-779d325fff6db493", "meaning-doc-53fcef4b84fc51e4", "meaning-doc-7bd9f95c76e3a5c9"], "hard_negative_texts": ["Ember payments memo 224: The approval threshold is capped at 48 days during Q2, after the first successful retry. Archived exceptions are counted as active for audit sampling.", "Ember payments memo 224: The approval threshold is capped at 51 days during Q2, when the customer tier is enterprise. The note also says archived exceptions remain active.", "Ember payments memo 224: The approval threshold target is at least 48 days during Q2, when the customer tier is enterprise. Teams may exceed it when capacity is low."], "id": "meaning-q-53652cd449131b1f", "positive_doc_id": "meaning-doc-8b654ee316e21aa8", "positive_text": "Ember payments memo 224: The approval threshold is capped at 48 days during Q2, when the customer tier is enterprise. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Ember's payments approval threshold has a 48 days cap in Q2 when the customer tier is enterprise, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "18 days; unless the owner is the platform team; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-0d13aed63173a4d2", "meaning-doc-80d87e17b016e028", "meaning-doc-8a645c815d7f616a"], "hard_negative_texts": ["Basil procurement memo 225: The review level is capped at 18 days during Q4, after the regional handoff is signed. Archived exceptions are counted as active for audit sampling.", "Basil procurement memo 225: The review level is capped at 16 days during Q4, unless the owner is the platform team. The note also says archived exceptions remain active.", "Basil procurement memo 225: The review level target is at least 18 days during Q4, unless the owner is the platform team. Teams may exceed it when capacity is low."], "id": "meaning-q-4f1675db56710d0b", "positive_doc_id": "meaning-doc-c52a02f7873ffed4", "positive_text": "Basil procurement memo 225: The review level is capped at 18 days during Q4, unless the owner is the platform team. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Basil's procurement review level has a 18 days cap in Q4 unless the owner is the platform team, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "36 minutes; unless the owner is the platform team; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-c779fd892396c421", "meaning-doc-7f02e92a770f4965", "meaning-doc-347bb69e8be73cdc"], "hard_negative_texts": ["Garnet procurement memo 226: The error budget is capped at 36 minutes during Q2, for every request regardless of source. Archived exceptions are counted as active for audit sampling.", "Garnet procurement memo 226: The error budget is capped at 38 minutes during Q2, unless the owner is the platform team. The note also says archived exceptions remain active.", "Garnet procurement memo 226: The error budget target is at least 36 minutes during Q2, unless the owner is the platform team. Teams may exceed it when capacity is low."], "id": "meaning-q-d7757918fcdc68fd", "positive_doc_id": "meaning-doc-c7428ca5f594a957", "positive_text": "Garnet procurement memo 226: The error budget is capped at 36 minutes during Q2, unless the owner is the platform team. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Garnet's procurement error budget has a 36 minutes cap in Q2 unless the owner is the platform team, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "36 percent; after the second failed retry; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-0567d4b6780ecb23", "meaning-doc-5083c2261b2b2439", "meaning-doc-463362914014c493"], "hard_negative_texts": ["Garnet privacy requests memo 227: The error budget is capped at 36 percent during Q1, only when the owner is the platform team. Archived exceptions are counted as active for audit sampling.", "Garnet privacy requests memo 227: The error budget is capped at 34 percent during Q1, after the second failed retry. The note also says archived exceptions remain active.", "Garnet privacy requests memo 227: The error budget target is at least 36 percent during Q1, after the second failed retry. Teams may exceed it when capacity is low."], "id": "meaning-q-67c842801b30f5b0", "positive_doc_id": "meaning-doc-c19d940da1143709", "positive_text": "Garnet privacy requests memo 227: The error budget is capped at 36 percent during Q1, after the second failed retry. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Garnet's privacy requests error budget has a 36 percent cap in Q1 after the second failed retry, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "24 minutes; unless the owner is the platform team; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-b94598149aaedcd6", "meaning-doc-29df09403c635f76", "meaning-doc-496beb49720b78db"], "hard_negative_texts": ["Fjord security review memo 228: The error budget is capped at 24 minutes during Q2, for every request regardless of source. Archived exceptions are counted as active for audit sampling.", "Fjord security review memo 228: The error budget is capped at 22 minutes during Q2, unless the owner is the platform team. The note also says archived exceptions remain active.", "Fjord security review memo 228: The error budget target is at least 24 minutes during Q2, unless the owner is the platform team. Teams may exceed it when capacity is low."], "id": "meaning-q-cb0b6bf089dff56b", "positive_doc_id": "meaning-doc-77d0235c291fb1fa", "positive_text": "Fjord security review memo 228: The error budget is capped at 24 minutes during Q2, unless the owner is the platform team. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Fjord's security review error budget has a 24 minutes cap in Q2 unless the owner is the platform team, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "12 percent; unless the owner is the platform team; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-69e1501eac14d004", "meaning-doc-9d51384f1b0cb12b", "meaning-doc-ce1e429da4f2faad"], "hard_negative_texts": ["Harbor warehouse operations memo 229: The latency is capped at 12 percent during Q3, after the regional handoff is signed. Archived exceptions are counted as active for audit sampling.", "Harbor warehouse operations memo 229: The latency is capped at 14 percent during Q3, unless the owner is the platform team. The note also says archived exceptions remain active.", "Harbor warehouse operations memo 229: The latency target is at least 12 percent during Q3, unless the owner is the platform team. Teams may exceed it when capacity is low."], "id": "meaning-q-29a985e588d07e90", "positive_doc_id": "meaning-doc-aefddc5aef9f4213", "positive_text": "Harbor warehouse operations memo 229: The latency is capped at 12 percent during Q3, unless the owner is the platform team. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Harbor's warehouse operations latency has a 12 percent cap in Q3 unless the owner is the platform team, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "12 tickets; unless the owner is the platform team; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-5c4fe0494ed45613", "meaning-doc-f9c574063454ec2f", "meaning-doc-6892c9982fcd2c0b"], "hard_negative_texts": ["Cedar privacy requests memo 230: The approval threshold is capped at 12 tickets during Q2, after the regional handoff is signed. Archived exceptions are counted as active for audit sampling.", "Cedar privacy requests memo 230: The approval threshold is capped at 9 tickets during Q2, unless the owner is the platform team. The note also says archived exceptions remain active.", "Cedar privacy requests memo 230: The approval threshold target is at least 12 tickets during Q2, unless the owner is the platform team. Teams may exceed it when capacity is low."], "id": "meaning-q-4f93b4ea969c27dc", "positive_doc_id": "meaning-doc-7bc1a37041720986", "positive_text": "Cedar privacy requests memo 230: The approval threshold is capped at 12 tickets during Q2, unless the owner is the platform team. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Cedar's privacy requests approval threshold has a 12 tickets cap in Q2 unless the owner is the platform team, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "12 percent; after the second failed retry; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-1a62796edf050ca3", "meaning-doc-0115f85f00d226ea", "meaning-doc-cbb3c898f5efa846"], "hard_negative_texts": ["Basil payments memo 231: The payment hold is capped at 12 percent during Q4, for every request regardless of source. Archived exceptions are counted as active for audit sampling.", "Basil payments memo 231: The payment hold is capped at 15 percent during Q4, after the second failed retry. The note also says archived exceptions remain active.", "Basil payments memo 231: The payment hold target is at least 12 percent during Q4, after the second failed retry. Teams may exceed it when capacity is low."], "id": "meaning-q-13769410034542ea", "positive_doc_id": "meaning-doc-71ad5f02309f4c73", "positive_text": "Basil payments memo 231: The payment hold is capped at 12 percent during Q4, after the second failed retry. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Basil's payments payment hold has a 12 percent cap in Q4 after the second failed retry, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "7 hours; unless the owner is the platform team; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-986146daa7a2f813", "meaning-doc-2c59377b81710b45", "meaning-doc-fa7f4cece725e6cb"], "hard_negative_texts": ["Kepler model deployment memo 232: The error budget is capped at 7 hours during Q4, after the regional handoff is signed. Archived exceptions are counted as active for audit sampling.", "Kepler model deployment memo 232: The error budget is capped at 10 hours during Q4, unless the owner is the platform team. The note also says archived exceptions remain active.", "Kepler model deployment memo 232: The error budget target is at least 7 hours during Q4, unless the owner is the platform team. Teams may exceed it when capacity is low."], "id": "meaning-q-108a81c4455b35dd", "positive_doc_id": "meaning-doc-69de3917086d653d", "positive_text": "Kepler model deployment memo 232: The error budget is capped at 7 hours during Q4, unless the owner is the platform team. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Kepler's model deployment error budget has a 7 hours cap in Q4 unless the owner is the platform team, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "48 days; only when the request is marked external; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-ae22da1c0985f4a0", "meaning-doc-55ea86324782d93d", "meaning-doc-7b44f23bc19b4c9f"], "hard_negative_texts": ["Kepler cloud billing memo 233: The latency is capped at 48 days during Q1, only when the owner is the platform team. Archived exceptions are counted as active for audit sampling.", "Kepler cloud billing memo 233: The latency is capped at 45 days during Q1, only when the request is marked external. The note also says archived exceptions remain active.", "Kepler cloud billing memo 233: The latency target is at least 48 days during Q1, only when the request is marked external. Teams may exceed it when capacity is low."], "id": "meaning-q-a6be32b0daf969bb", "positive_doc_id": "meaning-doc-269710e8b34ce32a", "positive_text": "Kepler cloud billing memo 233: The latency is capped at 48 days during Q1, only when the request is marked external. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Kepler's cloud billing latency has a 48 days cap in Q1 only when the request is marked external, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "72 percent; unless the owner is the platform team; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-3f9590fdae8b12fd", "meaning-doc-45586a85e0e3be94", "meaning-doc-877b0e601c15c3ab"], "hard_negative_texts": ["Juniper incident response memo 234: The approval threshold is capped at 72 percent during Q3, only when the owner is the platform team. Archived exceptions are counted as active for audit sampling.", "Juniper incident response memo 234: The approval threshold is capped at 78 percent during Q3, unless the owner is the platform team. The note also says archived exceptions remain active.", "Juniper incident response memo 234: The approval threshold target is at least 72 percent during Q3, unless the owner is the platform team. Teams may exceed it when capacity is low."], "id": "meaning-q-b398289d338451c8", "positive_doc_id": "meaning-doc-6b7a328ea4d8e41c", "positive_text": "Juniper incident response memo 234: The approval threshold is capped at 72 percent during Q3, unless the owner is the platform team. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Juniper's incident response approval threshold has a 72 percent cap in Q3 unless the owner is the platform team, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "12 hours; before the regional handoff is signed; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-5ab05339a2f85f21", "meaning-doc-fbd6ef7c2cdbce17", "meaning-doc-efc8fecae11a6db7"], "hard_negative_texts": ["Fjord clinical triage memo 235: The review level is capped at 12 hours during Q2, when the customer tier is starter. Archived exceptions are counted as active for audit sampling.", "Fjord clinical triage memo 235: The review level is capped at 15 hours during Q2, before the regional handoff is signed. The note also says archived exceptions remain active.", "Fjord clinical triage memo 235: The review level target is at least 12 hours during Q2, before the regional handoff is signed. Teams may exceed it when capacity is low."], "id": "meaning-q-cb7e31f8d56b491e", "positive_doc_id": "meaning-doc-e41b192cc6e2e000", "positive_text": "Fjord clinical triage memo 235: The review level is capped at 12 hours during Q2, before the regional handoff is signed. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Fjord's clinical triage review level has a 12 hours cap in Q2 before the regional handoff is signed, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "12 tickets; before the regional handoff is signed; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-2057a3a5b598db97", "meaning-doc-a664cda149369b1d", "meaning-doc-4bc0305f7b30a428"], "hard_negative_texts": ["Delta cloud billing memo 236: The sampling rate is capped at 12 tickets during Q4, after the regional handoff is signed. Archived exceptions are counted as active for audit sampling.", "Delta cloud billing memo 236: The sampling rate is capped at 9 tickets during Q4, before the regional handoff is signed. The note also says archived exceptions remain active.", "Delta cloud billing memo 236: The sampling rate target is at least 12 tickets during Q4, before the regional handoff is signed. Teams may exceed it when capacity is low."], "id": "meaning-q-f383eed18e723d79", "positive_doc_id": "meaning-doc-8fe76f0ee6125096", "positive_text": "Delta cloud billing memo 236: The sampling rate is capped at 12 tickets during Q4, before the regional handoff is signed. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Delta's cloud billing sampling rate has a 12 tickets cap in Q4 before the regional handoff is signed, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "18 minutes; unless the owner is the platform team; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-1a574a99e8bf7c86", "meaning-doc-10ffe118e3e56081", "meaning-doc-b6ee982c24d0bbab"], "hard_negative_texts": ["Juniper security review memo 237: The payment hold is capped at 18 minutes during Q1, after the regional handoff is signed. Archived exceptions are counted as active for audit sampling.", "Juniper security review memo 237: The payment hold is capped at 16 minutes during Q1, unless the owner is the platform team. The note also says archived exceptions remain active.", "Juniper security review memo 237: The payment hold target is at least 18 minutes during Q1, unless the owner is the platform team. Teams may exceed it when capacity is low."], "id": "meaning-q-5863ab013c7259f4", "positive_doc_id": "meaning-doc-dc3f082d7aae8264", "positive_text": "Juniper security review memo 237: The payment hold is capped at 18 minutes during Q1, unless the owner is the platform team. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Juniper's security review payment hold has a 18 minutes cap in Q1 unless the owner is the platform team, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "36 requests; unless the owner is the platform team; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-65088cbafaf7648a", "meaning-doc-c3a7a4e04a0e512d", "meaning-doc-8cc5142c0493e401"], "hard_negative_texts": ["Delta privacy requests memo 238: The sampling rate is capped at 36 requests during Q1, when the customer tier is starter. Archived exceptions are counted as active for audit sampling.", "Delta privacy requests memo 238: The sampling rate is capped at 34 requests during Q1, unless the owner is the platform team. The note also says archived exceptions remain active.", "Delta privacy requests memo 238: The sampling rate target is at least 36 requests during Q1, unless the owner is the platform team. Teams may exceed it when capacity is low."], "id": "meaning-q-7fba722f31539c53", "positive_doc_id": "meaning-doc-64e54b432a4b3ec1", "positive_text": "Delta privacy requests memo 238: The sampling rate is capped at 36 requests during Q1, unless the owner is the platform team. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Delta's privacy requests sampling rate has a 36 requests cap in Q1 unless the owner is the platform team, while archived exceptions are excluded.", "style": "meaning"} | |
| {"answer": "72 days; only when the request is marked external; archived exceptions excluded", "difficulty": "hard", "hard_negative_doc_ids": ["meaning-doc-4bc5a1d2c789f287", "meaning-doc-0a2ebc04c193a32f", "meaning-doc-fa1028fab61e461d"], "hard_negative_texts": ["Aurora model deployment memo 239: The rollback window is capped at 72 days during Q3, when the customer tier is starter. Archived exceptions are counted as active for audit sampling.", "Aurora model deployment memo 239: The rollback window is capped at 74 days during Q3, only when the request is marked external. The note also says archived exceptions remain active.", "Aurora model deployment memo 239: The rollback window target is at least 72 days during Q3, only when the request is marked external. Teams may exceed it when capacity is low."], "id": "meaning-q-5e5fb8cd36397fcf", "positive_doc_id": "meaning-doc-8e10e02838dd94a6", "positive_text": "Aurora model deployment memo 239: The rollback window is capped at 72 days during Q3, only when the request is marked external. The cap does not apply outside this scope, and archived exceptions must not be counted as active exceptions.", "query": "Find the memo where Aurora's model deployment rollback window has a 72 days cap in Q3 only when the request is marked external, while archived exceptions are excluded.", "style": "meaning"} | |