{ "checkpoint": "runs\\sovereign_specialists\\q-retriever-50m-v10\\final.pt", "holdout": "data\\model_suite\\q_retriever\\holdout\\q_retriever_v7_holdout.jsonl", "scorer": "exact_match", "n": 30, "mean": 1.0, "ci_lower": 1.0, "per_row": [ { "i": 0, "score": 1.0, "user": "Q: capital of Germany\n\nPASSAGE:\nBerlin is the capital of Germany.", "pred": "1", "gold": "1" }, { "i": 1, "score": 1.0, "user": "Q: capital of Germany\n\nPASSAGE:\nParis is the capital of France.", "pred": "0", "gold": "0" }, { "i": 2, "score": 1.0, "user": "Q: freezing point of water\n\nPASSAGE:\nWater freezes at 0 degrees Celsius.", "pred": "1", "gold": "1" }, { "i": 3, "score": 1.0, "user": "Q: freezing point of water\n\nPASSAGE:\nWater boils at 100 C.", "pred": "0", "gold": "0" }, { "i": 4, "score": 1.0, "user": "Q: net income amount\n\nPASSAGE:\nNet income for the period was $12 billion, up from $11 billion prior year.", "pred": "1", "gold": "1" }, { "i": 5, "score": 1.0, "user": "Q: net income amount\n\nPASSAGE:\nTotal revenue was $54 billion in the quarter.", "pred": "0", "gold": "0" }, { "i": 6, "score": 1.0, "user": "Q: how do I list files in a directory in python\n\nPASSAGE:\nUse `os.listdir(path)` or `pathlib.Path(path).iterdir()` to li", "pred": "1", "gold": "1" }, { "i": 7, "score": 1.0, "user": "Q: how do I list files in a directory in python\n\nPASSAGE:\nls is the Unix command.", "pred": "0", "gold": "0" }, { "i": 8, "score": 1.0, "user": "Q: what was free cash flow\n\nPASSAGE:\nFree cash flow for the year was 30 billion dollars.", "pred": "1", "gold": "1" }, { "i": 9, "score": 1.0, "user": "Q: what was free cash flow\n\nPASSAGE:\nOperating cash flow was $35 billion.", "pred": "0", "gold": "0" }, { "i": 10, "score": 1.0, "user": "Q: Python venv\n\nPASSAGE:\nUse `python -m venv .venv` to create a new isolated Python environment.", "pred": "1", "gold": "1" }, { "i": 11, "score": 1.0, "user": "Q: Python venv\n\nPASSAGE:\n`conda create -n myenv` makes a Conda environment.", "pred": "0", "gold": "0" }, { "i": 12, "score": 1.0, "user": "Q: capital of Germany\n\nPASSAGE:\nFree cash flow for the year was $30 billion.", "pred": "0", "gold": "0" }, { "i": 13, "score": 1.0, "user": "Q: capital of Germany\n\nPASSAGE:\nUse `pip install ` to install.", "pred": "0", "gold": "0" }, { "i": 14, "score": 1.0, "user": "Q: what was net income\n\nPASSAGE:\nBerlin is the capital of Germany.", "pred": "0", "gold": "0" }, { "i": 15, "score": 1.0, "user": "Q: what was net income\n\nPASSAGE:\nThe speed of light is about 3e8 m/s.", "pred": "0", "gold": "0" }, { "i": 16, "score": 1.0, "user": "Q: how to git commit\n\nPASSAGE:\nThe Nile is the world's longest river.", "pred": "0", "gold": "0" }, { "i": 17, "score": 1.0, "user": "Q: speed of light\n\nPASSAGE:\nThe speed of light in vacuum is approximately 299,792,458 meters per second.", "pred": "1", "gold": "1" }, { "i": 18, "score": 1.0, "user": "Q: speed of light\n\nPASSAGE:\nThe speed of sound in air is about 343 m/s.", "pred": "0", "gold": "0" }, { "i": 19, "score": 1.0, "user": "Q: Pythagorean theorem\n\nPASSAGE:\nThe Pythagorean theorem states that a^2 + b^2 = c^2 for a right triangle's sides.", "pred": "1", "gold": "1" }, { "i": 20, "score": 1.0, "user": "Q: Pythagorean theorem\n\nPASSAGE:\nHeron's formula gives area from side lengths.", "pred": "0", "gold": "0" }, { "i": 21, "score": 1.0, "user": "Q: when did World War II end\n\nPASSAGE:\nWorld War II ended in 1945, with Japan's surrender in September.", "pred": "1", "gold": "1" }, { "i": 22, "score": 1.0, "user": "Q: when did World War II end\n\nPASSAGE:\nWorld War I ended in 1918.", "pred": "0", "gold": "0" }, { "i": 23, "score": 1.0, "user": "Q: capital of Japan\n\nPASSAGE:\nTokyo is the capital of Japan.", "pred": "1", "gold": "1" }, { "i": 24, "score": 1.0, "user": "Q: capital of Japan\n\nPASSAGE:\nKyoto was the former capital of Japan before 1868.", "pred": "0", "gold": "0" }, { "i": 25, "score": 1.0, "user": "Q: CPR compression rate\n\nPASSAGE:\nPerform chest compressions at 100-120 per minute, about 2 inches deep.", "pred": "1", "gold": "1" }, { "i": 26, "score": 1.0, "user": "Q: CPR compression rate\n\nPASSAGE:\nHeimlich maneuver is for choking, not cardiac arrest.", "pred": "0", "gold": "0" }, { "i": 27, "score": 1.0, "user": "Q: EPS\n\nPASSAGE:\nDiluted EPS for the quarter was $2.40.", "pred": "1", "gold": "1" }, { "i": 28, "score": 1.0, "user": "Q: EPS\n\nPASSAGE:\nDividend per share was $1.20.", "pred": "0", "gold": "0" }, { "i": 29, "score": 1.0, "user": "Q: how to install Python on Windows\n\nPASSAGE:\nTo install Node.js on Windows, download the LTS installer from nodejs.org.", "pred": "0", "gold": "0" } ] }