Datasetpublic

chengfu0118/DeepSeek-R1-FP4_1757553968_eval_f912

chengfu0118/DeepSeek-R1-FP4_1757553968_eval_f912 Precomputed model outputs for evaluation. Evaluation Results GPQADiamond Average Accuracy: 61.62% ± 0.86% Number of Runs: 3 Run Accuracy Questions Solved Total Questions 1 62.12% 123 198 2 59.60% 118 198 3 63.13% 125 198

sourceHugging Faceupdated 1y agoView on Hugging Face

0likes6downloads

Dataset Card

chengfu0118/DeepSeek-R1-FP41757553968eval_f912

Precomputed model outputs for evaluation.

Evaluation Results

GPQADiamond

—Average Accuracy: 61.62% ± 0.86%
—Number of Runs: 3

Run	Accuracy	Questions Solved	Total Questions
1	62.12%	123	198
2	59.60%	118	198
3	63.13%	125	198