Spaces:
Running
Running
File size: 31,040 Bytes
c27a19a c1bb907 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 | <!DOCTYPE html>
<html lang="en">
<head>
<meta charset="utf-8" />
<meta name="viewport" content="width=device-width, initial-scale=1" />
<title>QuorumGuard & QuorumBench Explorer</title>
<style>
:root { --bg: #0f172a; --card: #1e293b; --border: #334155; --text: #f8fafc; --muted: #94a3b8; --accent: #10b981; --warn: #ef4444; }
body { font-family: system-ui, -apple-system, sans-serif; background: var(--bg); color: var(--text); margin: 0; padding: 24px; line-height: 1.5; }
.wrap { max-width: 1080px; margin: 0 auto; }
h1 { margin: 0 0 6px; font-size: 1.65rem; }
.sub { color: var(--muted); margin-bottom: 20px; font-size: 0.95rem; }
.sub a { color: #38bdf8; text-decoration: none; }
.grid { display: grid; grid-template-columns: repeat(auto-fit, minmax(310px, 1fr)); gap: 16px; margin-bottom: 20px; }
.card { background: var(--card); border: 1px solid var(--border); border-radius: 10px; padding: 16px; }
.card h2 { margin: 0 0 10px; font-size: 1.08rem; }
label { display: flex; justify-content: space-between; font-size: 0.85rem; color: var(--muted); margin-top: 8px; }
input[type=range], select { width: 100%; margin-top: 4px; }
select { background: #0f172a; color: var(--text); border: 1px solid var(--border); padding: 7px; border-radius: 6px; }
.stat { font-size: 1.25rem; font-weight: 700; color: var(--accent); }
.bad { color: var(--warn); }
table { width: 100%; border-collapse: collapse; font-size: 0.86rem; margin-top: 8px; }
th, td { text-align: left; padding: 7px 8px; border-bottom: 1px solid var(--border); }
th { color: var(--muted); font-weight: 600; }
.badge { padding: 2px 7px; border-radius: 4px; font-size: 0.76rem; font-weight: 600; }
.ok { background: rgba(16,185,129,0.18); color: #34d399; }
.err { background: rgba(239,68,68,0.18); color: #f87171; }
</style>
</head>
<body>
<div class="wrap">
<h1>QuorumGuard & QuorumBench Interactive Explorer</h1>
<div class="sub">
Live MPS Telemetry across Qwen2.5-3B, 7B, and 14B-Coder ·
<a href="https://huggingface.co/datasets/batistado/QuorumBench" target="_blank">Dataset</a> ·
<a href="https://huggingface.co/batistado/QuorumGuard" target="_blank">Hub Package</a> ·
<a href="https://github.com/batistado/quorumguard" target="_blank">GitHub</a>
</div>
<div class="grid">
<div class="card">
<h2>Theorem 1 Phase-Transition Calculator</h2>
<label><span>Shallow Error p0</span><span id="v_p0">0.78</span></label>
<input type="range" id="p0" min="0.52" max="0.95" step="0.01" value="0.78" />
<label><span>Asymptotic Deep Error p_inf</span><span id="v_pinf">0.08</span></label>
<input type="range" id="pinf" min="0.01" max="0.30" step="0.01" value="0.08" />
<label><span>Verification Rate lambda / mu</span><span id="v_rat">1.22</span></label>
<input type="range" id="rat" min="0.30" max="3.00" step="0.05" value="1.22" />
<label><span>Fast Quorum Ratio phi = k / N</span><span id="v_phi">0.35</span></label>
<input type="range" id="phi" min="0.10" max="0.99" step="0.01" value="0.35" />
<div style="margin-top:12px; font-size:0.88rem;">
Population Mean Error p_bar: <strong id="out_pbar"></strong><br/>
Critical Quorum Threshold phi_crit: <strong id="out_crit"></strong><br/>
Fast-Quorum Member Error p_fast(phi): <span id="out_pfast" class="stat"></span>
</div>
</div>
<div class="card">
<h2>QuorumBench-50 Summary (3B + 7B + 14B-Coder)</h2>
<table>
<thead><tr><th>Policy</th><th>Joint Acc</th><th>Trap Acc</th><th>P95 (s)</th></tr></thead>
<tbody>
<tr><td>Fast k-of-N</td><td class="bad">28.0%</td><td class="bad">6.7%</td><td>1.46s</td></tr>
<tr><td>CVaR Cutoff</td><td class="bad">28.0%</td><td class="bad">6.7%</td><td>5.60s</td></tr>
<tr><td>Sync N-of-N</td><td>42.0%</td><td>23.3%</td><td>49.86s</td></tr>
<tr><td>Genealogy Sync</td><td>72.0%</td><td>63.3%</td><td>48.86s</td></tr>
<tr><td><strong>QuorumGuard</strong></td><td><strong style="color:#34d399">72.0%</strong></td><td><strong style="color:#34d399">63.3%</strong></td><td><strong style="color:#34d399">14.82s</strong></td></tr>
</tbody>
</table>
</div>
</div>
<div class="card">
<h2>50-Workflow Live MPS Trace Browser</h2>
<select id="task_sel"></select>
<div id="task_detail" style="margin-top:12px; font-size:0.9rem;"></div>
</div>
</div>
<script>
const DATA = {"summary_50": {"Fast_k_of_N": {"stage1_accuracy": 0.46, "stage2_accuracy": 0.56, "end_to_end_joint_accuracy": 0.28, "boundary_trap_joint_accuracy": 0.0667, "routine_joint_accuracy": 0.6, "mean_workflow_latency_s": 1.1146, "routine_mean_latency_s": 1.0228, "p50_workflow_latency_s": 1.0775, "p95_workflow_latency_s": 1.459, "p99_workflow_latency_s": 1.5027, "cvar95_workflow_latency_s": 1.4891, "p_stage2_err_given_stage1_err": 0.4815, "p_stage2_err_given_stage1_ok": 0.3913, "stage1_boost_rate": 0.0}, "Sync_N_of_N": {"stage1_accuracy": 0.66, "stage2_accuracy": 0.68, "end_to_end_joint_accuracy": 0.42, "boundary_trap_joint_accuracy": 0.2333, "routine_joint_accuracy": 0.7, "mean_workflow_latency_s": 36.3392, "routine_mean_latency_s": 37.452, "p50_workflow_latency_s": 35.3277, "p95_workflow_latency_s": 49.8562, "p99_workflow_latency_s": 52.0719, "cvar95_workflow_latency_s": 51.4032, "p_stage2_err_given_stage1_err": 0.2353, "p_stage2_err_given_stage1_ok": 0.3636, "stage1_boost_rate": 0.0}, "Genealogy_Sync_HPR": {"stage1_accuracy": 0.84, "stage2_accuracy": 0.88, "end_to_end_joint_accuracy": 0.72, "boundary_trap_joint_accuracy": 0.6333, "routine_joint_accuracy": 0.85, "mean_workflow_latency_s": 35.9742, "routine_mean_latency_s": 37.452, "p50_workflow_latency_s": 35.1347, "p95_workflow_latency_s": 48.8589, "p99_workflow_latency_s": 52.0719, "cvar95_workflow_latency_s": 51.1952, "p_stage2_err_given_stage1_err": 0.0, "p_stage2_err_given_stage1_ok": 0.1429, "stage1_boost_rate": 0.0}, "CVaR_Static_Cutoff": {"stage1_accuracy": 0.46, "stage2_accuracy": 0.52, "end_to_end_joint_accuracy": 0.28, "boundary_trap_joint_accuracy": 0.0667, "routine_joint_accuracy": 0.6, "mean_workflow_latency_s": 5.6, "routine_mean_latency_s": 5.6, "p50_workflow_latency_s": 5.6, "p95_workflow_latency_s": 5.6, "p99_workflow_latency_s": 5.6, "cvar95_workflow_latency_s": 5.6, "p_stage2_err_given_stage1_err": 0.5556, "p_stage2_err_given_stage1_ok": 0.3913, "stage1_boost_rate": 0.0}, "QuorumGuard": {"stage1_accuracy": 0.84, "stage2_accuracy": 0.88, "end_to_end_joint_accuracy": 0.72, "boundary_trap_joint_accuracy": 0.6333, "routine_joint_accuracy": 0.85, "mean_workflow_latency_s": 10.1032, "routine_mean_latency_s": 8.4953, "p50_workflow_latency_s": 9.7349, "p95_workflow_latency_s": 14.8164, "p99_workflow_latency_s": 16.7789, "cvar95_workflow_latency_s": 16.1095, "p_stage2_err_given_stage1_err": 0.0, "p_stage2_err_given_stage1_ok": 0.1429, "stage1_boost_rate": 1.0}}, "tiers_50": {"fast_3b": {"model_tier": "Qwen2.5-3B Direct", "count": 150, "overall_accuracy": 0.5067, "boundary_trap_accuracy": 0.1889, "mean_completion_s": 0.5465, "mean_mps_wall_s": 0.0966, "mean_normalized_entropy": 0.2114}, "short_cot_3b": {"model_tier": "Qwen2.5-3B Short-CoT", "count": 50, "overall_accuracy": 0.4, "boundary_trap_accuracy": 0.2, "mean_completion_s": 2.691, "mean_mps_wall_s": 1.7752, "mean_normalized_entropy": 0.1873}, "cot_7b": {"model_tier": "Qwen2.5-7B CoT", "count": 50, "overall_accuracy": 0.64, "boundary_trap_accuracy": 0.4333, "mean_completion_s": 6.9671, "mean_mps_wall_s": 5.4683, "mean_normalized_entropy": 0.0938}, "verifier_7b": {"model_tier": "Qwen2.5-Coder-14B + Sandbox", "count": 100, "overall_accuracy": 0.8, "boundary_trap_accuracy": 0.7167, "mean_completion_s": 18.9132, "mean_mps_wall_s": 12.9253, "mean_normalized_entropy": 0.0928}}, "tasks_preview": [{"id": "R01_sha256_bytes", "regime": "routine", "cat": "systems_basics", "q1": "How many raw bytes (not hex characters) are produced by a standard SHA-256 digest?", "gold1": "B", "fast_c": "B", "fast_t": 0.9539, "fast_ok": true, "sync_t": 53.8198, "sync_ok": true, "qg_c": "B", "qg_t": 5.9087, "qg_ok": true}, {"id": "T01_uint16_overflow", "regime": "boundary_trap", "cat": "systems_bugs", "q1": "In C/CUDA, a kernel computes `uint16_t total_tokens = batch_size * seq_len;` When `batch_size = 256` and `seq_len = 256`, what exact value is stored in `total_tokens`?", "gold1": "C", "fast_c": "A", "fast_t": 1.0042, "fast_ok": false, "sync_t": 35.7391, "sync_ok": false, "qg_c": "A", "qg_t": 9.7123, "qg_ok": false}, {"id": "T02_float32_precision_stall", "regime": "boundary_trap", "cat": "numerical", "q1": "A telemetry counter uses IEEE-754 single-precision `float32` (`np.float32`). Starting at `x = np.float32(16777216.0)` (`2**24`), the loop runs `x = np.float32(x + np.float32(1.0))` 100 times. What is the exact final value of `x`?", "gold1": "B", "fast_c": "A", "fast_t": 1.4622, "fast_ok": false, "sync_t": 42.9566, "sync_ok": true, "qg_c": "B", "qg_t": 14.4467, "qg_ok": true}, {"id": "R02_binary_heap_left_child", "regime": "routine", "cat": "algorithms", "q1": "In a 0-indexed binary heap stored in a contiguous array, what is the index of the left child of node i = 14?", "gold1": "C", "fast_c": "C", "fast_t": 0.9658, "fast_ok": false, "sync_t": 31.8221, "sync_ok": false, "qg_c": "C", "qg_t": 7.9728, "qg_ok": false}, {"id": "T03_sql_not_in_null", "regime": "boundary_trap", "cat": "databases", "q1": "Table `orders` has 5 rows with `id` in `(1, 2, 3, 4, 5)`. Subquery `blocked_ids` returns `(2, 4, NULL)`. How many rows does `SELECT id FROM orders WHERE id NOT IN (SELECT uid FROM blocked_ids);` return in standard SQL / PostgreSQL / SQLite?", "gold1": "B", "fast_c": "A", "fast_t": 0.7921, "fast_ok": false, "sync_t": 33.7614, "sync_ok": false, "qg_c": "A", "qg_t": 7.6593, "qg_ok": false}, {"id": "T04_shift_add_precedence", "regime": "boundary_trap", "cat": "systems_bugs", "q1": "In both C and Python 3, what exact integer value does the unparenthesized expression `1 << 2 + 3` evaluate to?", "gold1": "B", "fast_c": "A", "fast_t": 1.2246, "fast_ok": false, "sync_t": 30.8385, "sync_ok": true, "qg_c": "B", "qg_t": 8.3332, "qg_ok": true}, {"id": "R03_http_idempotent", "regime": "routine", "cat": "distributed_systems", "q1": "According to RFC 9110 (HTTP Semantics), which of the following HTTP methods is NOT guaranteed to be idempotent?", "gold1": "D", "fast_c": "D", "fast_t": 0.9549, "fast_ok": false, "sync_t": 27.9196, "sync_ok": true, "qg_c": "D", "qg_t": 7.0227, "qg_ok": true}, {"id": "T05_python_xor_caret", "regime": "boundary_trap", "cat": "systems_bugs", "q1": "In Python 3, what is the exact integer value evaluated by the expression `3 ^ 4 % 5`?", "gold1": "B", "fast_c": "A", "fast_t": 1.1115, "fast_ok": false, "sync_t": 31.746, "sync_ok": true, "qg_c": "B", "qg_t": 10.4643, "qg_ok": true}, {"id": "T06_python_bankers_rounding", "regime": "boundary_trap", "cat": "numerical", "q1": "In Python 3, evaluate `sum(round(x + 0.5) for x in range(8))` (i.e., `round(0.5) + round(1.5) + round(2.5) + round(3.5) + round(4.5) + round(5.5) + round(6.5) + round(7.5)`).", "gold1": "B", "fast_c": "A", "fast_t": 0.9381, "fast_ok": false, "sync_t": 31.526, "sync_ok": true, "qg_c": "B", "qg_t": 10.6459, "qg_ok": true}, {"id": "R04_matmul_flops", "regime": "routine", "cat": "numerical", "q1": "If matrix X has shape (4, 8) and matrix Y has shape (8, 5), what is the output shape of the matrix product X @ Y?", "gold1": "A", "fast_c": "A", "fast_t": 1.0227, "fast_ok": false, "sync_t": 37.1754, "sync_ok": true, "qg_c": "A", "qg_t": 7.6389, "qg_ok": true}, {"id": "T07_closure_late_binding", "regime": "boundary_trap", "cat": "systems_bugs", "q1": "What exact value is returned by the Python 3 snippet:\n`funcs = [lambda x: x * i for i in range(5)]\nresult = funcs[1](10) + funcs[2](10)`", "gold1": "B", "fast_c": "A", "fast_t": 1.2037, "fast_ok": false, "sync_t": 30.4771, "sync_ok": true, "qg_c": "B", "qg_t": 10.6556, "qg_ok": true}, {"id": "T08_shallow_list_multiply", "regime": "boundary_trap", "cat": "systems_bugs", "q1": "What exact integer is printed by this Python snippet?\n`dp = [[0] * 4] * 5\ndp[0][2] = 7\ndp[3][2] += 2\nprint(sum(row[2] for row in dp))`", "gold1": "B", "fast_c": "A", "fast_t": 1.1075, "fast_ok": false, "sync_t": 30.2633, "sync_ok": true, "qg_c": "B", "qg_t": 9.7897, "qg_ok": true}, {"id": "R05_tcp_syn_seq", "regime": "routine", "cat": "networking", "q1": "During a TCP 3-way handshake, if the client sends SYN with sequence number ISN = 1000, what acknowledgment number (ACK) does the server send in its SYN-ACK packet?", "gold1": "C", "fast_c": "C", "fast_t": 0.9959, "fast_ok": true, "sync_t": 48.0594, "sync_ok": true, "qg_c": "C", "qg_t": 8.2769, "qg_ok": true}, {"id": "T09_utf8_emoji_bytes", "regime": "boundary_trap", "cat": "systems_bugs", "q1": "In Python 3, what exact integer does `len('café 🚀'.encode('utf-8'))` return? (Note: string is `'c'`, `'a'`, `'f'`, `'é'` U+00E9, `' '` U+0020, `'🚀'` U+1F680.)", "gold1": "C", "fast_c": "B", "fast_t": 1.2334, "fast_ok": false, "sync_t": 32.3774, "sync_ok": true, "qg_c": "C", "qg_t": 9.2135, "qg_ok": true}, {"id": "T10_harmonic_throughput", "regime": "boundary_trap", "cat": "distributed_systems", "q1": "A backup job uploads a 600 GB snapshot over Link A at 60 MB/s, and immediately downloads/verifies the exact same 600 GB snapshot over Link B at 20 MB/s. What is the exact effective average transfer rate (total GB transferred divided by total elapsed time) across the round-trip?", "gold1": "B", "fast_c": "D", "fast_t": 1.3012, "fast_ok": false, "sync_t": 30.7587, "sync_ok": true, "qg_c": "B", "qg_t": 11.2849, "qg_ok": true}, {"id": "R06_raft_quorum_size", "regime": "routine", "cat": "distributed_systems", "q1": "In a standard symmetric Raft cluster of N = 9 voting nodes, what is the minimum number of nodes required to form a commit quorum?", "gold1": "B", "fast_c": "B", "fast_t": 1.0095, "fast_ok": false, "sync_t": 35.9513, "sync_ok": false, "qg_c": "B", "qg_t": 9.3677, "qg_ok": false}, {"id": "T11_bayes_false_positive", "regime": "boundary_trap", "cat": "math", "q1": "In a cluster of 10,000 nodes, exactly 0.1% (10 nodes) truly have silent hardware corruption. An automated detector flags corrupt nodes with 99% true-positive sensitivity, and has a 1% false-positive rate on healthy nodes (99% specificity). If a randomly chosen node triggers an alert, what is the closest posterior probability that the node truly has silent corruption?", "gold1": "C", "fast_c": "A", "fast_t": 0.9783, "fast_ok": false, "sync_t": 32.717, "sync_ok": false, "qg_c": "B", "qg_t": 11.79, "qg_ok": false}, {"id": "T12_int32_midpoint_overflow", "regime": "boundary_trap", "cat": "algorithms", "q1": "In C/Java (`int32_t`, signed 32-bit two's complement range `-2,147,483,648` to `2,147,483,647`), binary search computes `int32_t mid = (low + high) / 2;` with `low = 1500000000` and `high = 1700000000`. Wrapping in 32-bit two's complement before `/ 2`, what happens to `mid`?", "gold1": "B", "fast_c": "B", "fast_t": 1.1974, "fast_ok": true, "sync_t": 37.1413, "sync_ok": true, "qg_c": "B", "qg_t": 7.8384, "qg_ok": true}, {"id": "R07_dag_topo_cycle", "regime": "routine", "cat": "algorithms", "q1": "Kahn's topological sort algorithm runs on a directed graph with V = 12 vertices and terminates having popped only 9 vertices from the zero-indegree queue. What does this guarantee?", "gold1": "B", "fast_c": "B", "fast_t": 1.0934, "fast_ok": true, "sync_t": 27.9027, "sync_ok": true, "qg_c": "B", "qg_t": 6.6413, "qg_ok": true}, {"id": "T13_sliding_window_fencepost", "regime": "boundary_trap", "cat": "algorithms", "q1": "A function computes the maximum sum of any contiguous subarray of length `k = 4` over `arr = [2, 1, 5, 1, 3, 2, 9]` (`len(arr) = 7`) using:\n`best = max(sum(arr[i : i + k]) for i in range(len(arr) - k))`\nWhat exact value does `best` evaluate to?", "gold1": "B", "fast_c": "A", "fast_t": 1.2888, "fast_ok": false, "sync_t": 35.1901, "sync_ok": false, "qg_c": "B", "qg_t": 10.9806, "qg_ok": false}, {"id": "T14_mutable_default_retry", "regime": "boundary_trap", "cat": "systems_bugs", "q1": "Consider Python function:\n`def record_attempt(code, history=[]):\n history.append(code)\n return sum(history)`\nIf a worker calls `a = record_attempt(200)`, then `b = record_attempt(500)`, then `c = record_attempt(200)`, what is the exact value of `a + b + c`?", "gold1": "B", "fast_c": "A", "fast_t": 0.8843, "fast_ok": false, "sync_t": 31.4425, "sync_ok": true, "qg_c": "B", "qg_t": 11.5479, "qg_ok": true}, {"id": "R08_ipv4_cidr_hosts", "regime": "routine", "cat": "networking", "q1": "How many total IPv4 addresses (including network and broadcast addresses) are contained in a /26 CIDR block?", "gold1": "B", "fast_c": "B", "fast_t": 0.9297, "fast_ok": true, "sync_t": 31.0056, "sync_ok": true, "qg_c": "B", "qg_t": 7.5074, "qg_ok": true}, {"id": "T15_neg_floor_mod_c_vs_py", "regime": "boundary_trap", "cat": "numerical", "q1": "In Python 3, what exact tuple is produced by `(-17 // 5, -17 % 5)`?", "gold1": "B", "fast_c": "A", "fast_t": 1.394, "fast_ok": false, "sync_t": 39.0865, "sync_ok": false, "qg_c": "B", "qg_t": 8.1634, "qg_ok": false}, {"id": "T16_generator_exhaustion", "regime": "boundary_trap", "cat": "systems_bugs", "q1": "What exact float value is computed by this Python 3 snippet?\n`scores = (x * 2 for x in [1, 2, 3, 4])\ntotal = sum(scores)\ncount = len(list(scores))\navg = total / max(count, 1)`", "gold1": "B", "fast_c": "A", "fast_t": 1.0429, "fast_ok": false, "sync_t": 30.2101, "sync_ok": true, "qg_c": "B", "qg_t": 10.6851, "qg_ok": true}, {"id": "R09_btree_search_complexity", "regime": "routine", "cat": "databases", "q1": "What is the worst-case asymptotic time complexity of point-key lookup in a balanced B-Tree indexing N keys?", "gold1": "B", "fast_c": "B", "fast_t": 1.0357, "fast_ok": true, "sync_t": 31.2164, "sync_ok": true, "qg_c": "B", "qg_t": 7.1548, "qg_ok": true}, {"id": "T17_float_nan_sort_membership", "regime": "boundary_trap", "cat": "numerical", "q1": "In Python 3, execute:\n`v = float('nan')\na = (v == v)\nb = (v in [v])`\nWhat exact pair `(a, b)` is produced?", "gold1": "B", "fast_c": "A", "fast_t": 1.1835, "fast_ok": false, "sync_t": 35.4653, "sync_ok": true, "qg_c": "B", "qg_t": 11.5597, "qg_ok": true}, {"id": "T18_modulo_chain_remainder", "regime": "boundary_trap", "cat": "math", "q1": "What is the exact remainder when `7**2026 + 3**2026` is divided by `10`?", "gold1": "B", "fast_c": "A", "fast_t": 1.3003, "fast_ok": false, "sync_t": 32.1752, "sync_ok": false, "qg_c": "B", "qg_t": 10.8512, "qg_ok": false}, {"id": "R10_geometric_series", "regime": "routine", "cat": "math", "q1": "What is the exact sum of the finite geometric series 1 + 2 + 4 + 8 + 16 + 32 + 64 + 128?", "gold1": "B", "fast_c": "B", "fast_t": 1.0857, "fast_ok": true, "sync_t": 50.2527, "sync_ok": true, "qg_c": "B", "qg_t": 12.1493, "qg_ok": true}, {"id": "T19_sql_count_col_null", "regime": "boundary_trap", "cat": "databases", "q1": "Table `sessions` has 6 rows with column `latency_ms` values `(12, NULL, 18, NULL, 0, 25)`. What exact pair of integers does `SELECT COUNT(*), COUNT(latency_ms) FROM sessions;` return in SQL?", "gold1": "B", "fast_c": "B", "fast_t": 0.8291, "fast_ok": false, "sync_t": 38.1706, "sync_ok": true, "qg_c": "B", "qg_t": 14.1415, "qg_ok": true}, {"id": "T20_python_dict_bool_int_float_keys", "regime": "boundary_trap", "cat": "systems_bugs", "q1": "In Python 3, execute:\n`d = {True: 'yes', 1: 'int_one', 1.0: 'float_one'}\nprint(len(d), list(d.keys())[0], list(d.values())[0])`\nWhat exact output is printed?", "gold1": "B", "fast_c": "A", "fast_t": 1.2794, "fast_ok": false, "sync_t": 39.4202, "sync_ok": false, "qg_c": "A", "qg_t": 7.8341, "qg_ok": false}, {"id": "R11_mutex_deadlock_order", "regime": "routine", "cat": "concurrency", "q1": "Thread 1 locks Mutex A then Mutex B, while Thread 2 locks Mutex B then Mutex A. Which classic Coffman condition is violated if we enforce a strict global lock acquisition order (always lock A before B)?", "gold1": "B", "fast_c": "B", "fast_t": 0.9876, "fast_ok": true, "sync_t": 40.6185, "sync_ok": true, "qg_c": "B", "qg_t": 5.6695, "qg_ok": true}, {"id": "T21_signed_right_shift_mask", "regime": "boundary_trap", "cat": "systems_bugs", "q1": "In Python 3 (and two's-complement arithmetic right shift), what exact integer does `-1 >> 5` evaluate to?", "gold1": "B", "fast_c": "A", "fast_t": 1.0571, "fast_ok": false, "sync_t": 31.7615, "sync_ok": true, "qg_c": "B", "qg_t": 10.3015, "qg_ok": true}, {"id": "T22_dijkstra_negative_edge", "regime": "boundary_trap", "cat": "algorithms", "q1": "Directed graph has 4 nodes `{S, A, B, T}` and edges `S->A (weight 2)`, `S->B (weight 5)`, `B->A (weight -4)`, `A->T (weight 1)`. Standard non-revisiting Dijkstra (marking popped nodes finalized once) runs from `S`. What `(dijkstra_output, true_shortest_path)` pair for `S -> T` occurs?", "gold1": "B", "fast_c": "A", "fast_t": 1.2296, "fast_ok": false, "sync_t": 49.5131, "sync_ok": false, "qg_c": "A", "qg_t": 18.183, "qg_ok": false}, {"id": "R12_kv_cache_bytes", "regime": "routine", "cat": "ml_systems", "q1": "A transformer layer caches K and V tensors for 1 sequence of 1024 tokens with 8 KV heads, head_dim = 128, stored in FP16 (2 bytes/element). How many megabytes (1 MB = 1,048,576 bytes) does this single layer's KV cache occupy?", "gold1": "C", "fast_c": "C", "fast_t": 1.4979, "fast_ok": false, "sync_t": 44.154, "sync_ok": true, "qg_c": "C", "qg_t": 14.828, "qg_ok": true}, {"id": "T23_python_chained_comparison", "regime": "boundary_trap", "cat": "systems_bugs", "q1": "In Python 3, what exact boolean value does the expression `(False == False) in [False]` vs `False == False in [False]` evaluate to, specifically `False == False in [False]`?", "gold1": "B", "fast_c": "B", "fast_t": 1.2584, "fast_ok": true, "sync_t": 39.7571, "sync_ok": true, "qg_c": "B", "qg_t": 14.8023, "qg_ok": true}, {"id": "T24_c_sizeof_array_param_decay", "regime": "boundary_trap", "cat": "systems_bugs", "q1": "On 64-bit Linux (`LP64`, `sizeof(int) == 4`, `sizeof(int*) == 8`), a C function is defined as:\n`size_t get_len(int arr[100]) { return sizeof(arr) / sizeof(arr[0]); }`\nWhat exact value does `get_len(my_buf)` return?", "gold1": "B", "fast_c": "A", "fast_t": 1.1585, "fast_ok": false, "sync_t": 37.7457, "sync_ok": true, "qg_c": "B", "qg_t": 13.3228, "qg_ok": true}, {"id": "R13_http_429_status", "regime": "routine", "cat": "distributed_systems", "q1": "In standard HTTP status codes (RFC 6585), what condition does status code 429 indicate?", "gold1": "B", "fast_c": "B", "fast_t": 1.0424, "fast_ok": true, "sync_t": 40.5339, "sync_ok": true, "qg_c": "B", "qg_t": 8.9713, "qg_ok": true}, {"id": "T25_list_remove_while_iterating", "regime": "boundary_trap", "cat": "systems_bugs", "q1": "In Python 3, execute:\n`nums = [2, 4, 6, 8, 9]\nfor x in nums:\n if x % 2 == 0:\n nums.remove(x)`\nWhat is the exact final list `nums`?", "gold1": "B", "fast_c": "A", "fast_t": 1.5073, "fast_ok": false, "sync_t": 35.0141, "sync_ok": false, "qg_c": "D", "qg_t": 13.226, "qg_ok": false}, {"id": "T26_modular_inverse_existence", "regime": "boundary_trap", "cat": "math", "q1": "How many integers `x` in `{0, 1, ..., 20}` satisfy the modular congruence `(6 * x) % 21 == 1`?", "gold1": "B", "fast_c": "A", "fast_t": 1.2405, "fast_ok": false, "sync_t": 36.4605, "sync_ok": true, "qg_c": "B", "qg_t": 12.6878, "qg_ok": true}, {"id": "R14_quicksort_worst_case", "regime": "routine", "cat": "algorithms", "q1": "What is the worst-case time complexity of deterministic first-element-pivot Quicksort on an already sorted array of N distinct elements?", "gold1": "C", "fast_c": "C", "fast_t": 0.9869, "fast_ok": true, "sync_t": 37.89, "sync_ok": true, "qg_c": "C", "qg_t": 7.2944, "qg_ok": true}, {"id": "T27_python_string_strip_chars", "regime": "boundary_trap", "cat": "systems_bugs", "q1": "In Python 3, what exact string does `'prefix_price_id'.lstrip('prefix_')` return?", "gold1": "B", "fast_c": "A", "fast_t": 1.294, "fast_ok": false, "sync_t": 28.4513, "sync_ok": false, "qg_c": "A", "qg_t": 6.7987, "qg_ok": false}, {"id": "T28_amdahl_parallel_limit", "regime": "boundary_trap", "cat": "distributed_systems", "q1": "A workload has a strictly serial initialization/commit fraction of `s = 20%` (`0.20`) and a perfectly parallelizable fraction of `80%` (`0.80`). Under Amdahl's Law, if we scale from 1 GPU to `P = 16` GPUs, what is the exact end-to-end speedup (rounded to 2 decimal places)?", "gold1": "B", "fast_c": "D", "fast_t": 1.2458, "fast_ok": false, "sync_t": 36.9528, "sync_ok": true, "qg_c": "B", "qg_t": 13.9217, "qg_ok": true}, {"id": "R15_raid1_capacity", "regime": "routine", "cat": "systems_basics", "q1": "A storage server pairs two identical 4 TB NVMe drives in a standard RAID-1 (full mirroring) array. What is the usable logical storage capacity?", "gold1": "B", "fast_c": "B", "fast_t": 0.9523, "fast_ok": true, "sync_t": 28.2279, "sync_ok": true, "qg_c": "B", "qg_t": 7.6487, "qg_ok": true}, {"id": "T29_twos_complement_abs_int8_min", "regime": "boundary_trap", "cat": "numerical", "q1": "In 8-bit signed two's-complement arithmetic (`np.int8`, range `-128` to `127`), wrapping modulo 256 in signed `int8`, what value does `-np.int8(-128)` produce?", "gold1": "B", "fast_c": "D", "fast_t": 1.0692, "fast_ok": false, "sync_t": 26.748, "sync_ok": true, "qg_c": "B", "qg_t": 9.0963, "qg_ok": true}, {"id": "T30_lease_clock_skew_split_brain", "regime": "boundary_trap", "cat": "distributed_systems", "q1": "Leader L1 acquires a 10.0-second lease at true physical time `t = 0.0s`. L1's local monotonic clock runs `15%` slow (`0.85s` measured per `1.0s` real time), while the Lock Server's clock runs `5%` fast (`1.05s` per `1.0s` real time) and grants Leader L2 the lease as soon as `10.0s` elapse on the Lock Server's clock. For how many real physical seconds do L1 and L2 BOTH believe they hold the active exclusive lease simultaneously (split-brain window)?", "gold1": "B", "fast_c": "A", "fast_t": 1.455, "fast_ok": false, "sync_t": 45.805, "sync_ok": false, "qg_c": "A", "qg_t": 15.3175, "qg_ok": false}, {"id": "R16_bloom_filter_fn", "regime": "routine", "cat": "algorithms", "q1": "In a standard insert-only Bloom filter (without deletions or bit corruption), what is the exact theoretical false-negative rate for keys that were previously inserted?", "gold1": "A", "fast_c": "A", "fast_t": 0.936, "fast_ok": true, "sync_t": 40.8189, "sync_ok": true, "qg_c": "A", "qg_t": 9.4695, "qg_ok": true}, {"id": "R17_euler_dag_edges", "regime": "routine", "cat": "math", "q1": "How many edges does a connected undirected tree with V = 25 vertices contain?", "gold1": "B", "fast_c": "B", "fast_t": 0.9069, "fast_ok": false, "sync_t": 29.1548, "sync_ok": true, "qg_c": "B", "qg_t": 7.2512, "qg_ok": true}, {"id": "R18_page_table_offset_bits", "regime": "routine", "cat": "systems_basics", "q1": "On an architecture using 4 KiB (4096-byte) virtual memory pages, how many least-significant virtual address bits form the intra-page byte offset?", "gold1": "B", "fast_c": "B", "fast_t": 0.9934, "fast_ok": false, "sync_t": 44.6066, "sync_ok": false, "qg_c": "B", "qg_t": 12.9617, "qg_ok": false}, {"id": "R19_attention_mask_causal", "regime": "routine", "cat": "ml_systems", "q1": "In causal self-attention over a sequence of L = 10 tokens (0-indexed 0..9), how many non-masked key positions (including itself) can query token at index q = 6 attend to?", "gold1": "B", "fast_c": "B", "fast_t": 1.1746, "fast_ok": false, "sync_t": 35.0792, "sync_ok": true, "qg_c": "B", "qg_t": 9.7574, "qg_ok": true}, {"id": "R20_acid_isolation_serializable", "regime": "routine", "cat": "databases", "q1": "Among the four standard ANSI SQL transaction isolation levels, which is the strictest level that prevents dirty reads, non-repeatable reads, AND phantom reads?", "gold1": "D", "fast_c": "D", "fast_t": 0.9316, "fast_ok": true, "sync_t": 32.8308, "sync_ok": true, "qg_c": "D", "qg_t": 6.4139, "qg_ok": true}]};
function pFast(phi, p0, pinf, ratio) {
if (phi >= 0.9999) return pinf + (p0 - pinf) * (1.0 / (1.0 + ratio));
const exp = 1.0 + ratio;
const term = (1.0 / (1.0 + ratio)) * (1.0 - Math.pow(1.0 - phi, exp));
return pinf + (p0 - pinf) * (term / phi);
}
function updateCalc() {
const p0 = parseFloat(document.getElementById('p0').value);
const pinf = parseFloat(document.getElementById('pinf').value);
const rat = parseFloat(document.getElementById('rat').value);
const phi = parseFloat(document.getElementById('phi').value);
document.getElementById('v_p0').textContent = p0.toFixed(2);
document.getElementById('v_pinf').textContent = pinf.toFixed(2);
document.getElementById('v_rat').textContent = rat.toFixed(2);
document.getElementById('v_phi').textContent = phi.toFixed(2);
const pbar = pinf + (p0 - pinf) * (1.0 / (1.0 + rat));
let lo = 0.001, hi = 0.9999;
for (let i = 0; i < 50; i++) {
const mid = 0.5 * (lo + hi);
if (pFast(mid, p0, pinf, rat) > 0.5) lo = mid; else hi = mid;
}
const crit = (pbar >= 0.5) ? 1.0 : 0.5 * (lo + hi);
const pf = pFast(phi, p0, pinf, rat);
document.getElementById('out_pbar').textContent = pbar.toFixed(3) + (pbar < 0.5 ? ' (Sync -> 0% err)' : ' (Unreliable)');
document.getElementById('out_crit').textContent = crit.toFixed(3);
const el = document.getElementById('out_pfast');
el.textContent = pf.toFixed(3) + (pf > 0.5 ? ' -> FALSE CONSENSUS' : ' -> Stable Quorum');
el.className = pf > 0.5 ? 'stat bad' : 'stat';
}
['p0','pinf','rat','phi'].forEach(id => document.getElementById(id).addEventListener('input', updateCalc));
updateCalc();
const sel = document.getElementById('task_sel');
DATA.tasks_preview.forEach((t, idx) => {
const opt = document.createElement('option');
opt.value = idx;
opt.textContent = `[${t.regime.toUpperCase()}] ${t.id} (${t.cat})`;
sel.appendChild(opt);
});
function renderTask() {
const t = DATA.tasks_preview[parseInt(sel.value, 10)];
const b = ok => `<span class="badge ${ok ? 'ok' : 'err'}">${ok ? 'BOTH STAGES OK' : 'FAILED'}</span>`;
document.getElementById('task_detail').innerHTML = `
<div style="color:#cbd5e1; margin-bottom:8px;"><strong>Stage-1 Prompt:</strong> ${t.q1} (Gold: <code>${t.gold1}</code>)</div>
<table>
<thead><tr><th>Policy</th><th>Stage-1 Commit</th><th>2-Stage Latency</th><th>Workflow Outcome</th></tr></thead>
<tbody>
<tr><td>Fast k-of-N</td><td><code>${t.fast_c}</code></td><td>${t.fast_t.toFixed(2)}s</td><td>${b(t.fast_ok)}</td></tr>
<tr><td>Genealogy Sync (HPR)</td><td><code>${t.qg_c}</code></td><td>${t.sync_t.toFixed(2)}s</td><td>${b(t.sync_ok)}</td></tr>
<tr><td><strong>QuorumGuard</strong></td><td><code>${t.qg_c}</code></td><td><strong>${t.qg_t.toFixed(2)}s</strong></td><td>${b(t.qg_ok)}</td></tr>
</tbody>
</table>`;
}
sel.addEventListener('change', renderTask);
renderTask();
</script>
</body>
</html>
|