README.md

mlfoundations-dev/a1_code_octopack_sft_eval_636d

Precomputed model outputs for evaluation.

Evaluation Results

Summary

Metric	AIME24	AMC23	MATH500	MMLUPro	JEEBench	GPQADiamond	LiveCodeBench	CodeElo	CodeForces
Accuracy	16.0	53.2	71.6	26.4	37.5	41.8	25.3	5.1	6.1

AIME24

• Average Accuracy: 16.00% ± 1.03%
• Number of Runs: 10

Run	Accuracy	Questions Solved	Total Questions
1	16.67%	5	30
2	20.00%	6	30
3	16.67%	5	30
4	13.33%	4	30
5	10.00%	3	30
6	16.67%	5	30
7	20.00%	6	30
8	13.33%	4	30
9	20.00%	6	30
10	13.33%	4	30

AMC23

• Average Accuracy: 53.25% ± 1.42%
• Number of Runs: 10

Run	Accuracy	Questions Solved	Total Questions
1	50.00%	20	40
2	57.50%	23	40
3	57.50%	23	40
4	57.50%	23	40
5	57.50%	23	40
6	45.00%	18	40
7	55.00%	22	40
8	55.00%	22	40
9	47.50%	19	40
10	50.00%	20	40

MATH500

• Accuracy: 71.60%

  Accuracy	Questions Solved	Total Questions
  71.60%	358	500

MMLUPro

• Average Accuracy: 26.40% ± 0.00%
• Number of Runs: 1

Run	Accuracy	Questions Solved	Total Questions
1	26.40%	132	500

JEEBench

• Average Accuracy: 37.51% ± 0.82%
• Number of Runs: 3

Run	Accuracy	Questions Solved	Total Questions
1	36.17%	186.25	515
2	36.89%	190.0	515
3	39.47%	203.25	515

GPQADiamond

• Average Accuracy: 41.75% ± 2.96%
• Number of Runs: 3

Run	Accuracy	Questions Solved	Total Questions
1	37.88%	75	198
2	38.38%	76	198
3	48.99%	97	198

LiveCodeBench

• Average Accuracy: 25.31% ± 0.73%
• Number of Runs: 3

Run	Accuracy	Questions Solved	Total Questions
1	23.87%	122	511
2	25.83%	132	511
3	26.22%	134	511

CodeElo

• Average Accuracy: 5.12% ± 0.15%
• Number of Runs: 3

Run	Accuracy	Questions Solved	Total Questions
1	4.86%	19	391
2	5.37%	21	391
3	5.12%	20	391

CodeForces

• Average Accuracy: 6.11% ± 0.70%
• Number of Runs: 3

Run	Accuracy	Questions Solved	Total Questions
1	7.28%	33	453
2	6.18%	28	453
3	4.86%	22	453