kth8/gpt-oss-120b-SuperGPQA-benchmark
收藏资源简介:
--- license: apache-2.0 language: - en base_model: openai/gpt-oss-120b datasets: - m-a-p/SuperGPQA --- Benchmark of [openai/gpt-oss-120b](https://huggingface.co/openai/gpt-oss-120b) against [m-a-p/SuperGPQA](https://huggingface.co/datasets/m-a-p/SuperGPQA) dataset. Accuracy: 54.900000000000006% with Python tool. | Metric | Value | |----------------------|---------------| | **Correct** | 549 | | **Incorrect** | 448 | | **Errors** | 3 | | **Total samples** | 1000 | | **Python tool calls**| 889 | | **Total completion tokens** | 1,301,333 | Raw stats: ```json { "accuracy": 0.549, "correct": 549, "incorrect": 448, "error": 3, "total": 1000, "python_tool_calls": 889, "completion_tokens": 1301333 } ```
许可证:Apache-2.0 语言:英语 基础模型:openai/gpt-oss-120b 测试数据集:m-a-p/SuperGPQA 本基准测试基于[m-a-p/SuperGPQA](https://huggingface.co/datasets/m-a-p/SuperGPQA)数据集,对[openai/gpt-oss-120b](https://huggingface.co/openai/gpt-oss-120b)模型开展性能评测。经Python工具计算,模型准确率为54.900000000000006%。评测指标与对应数值如下表: | 评测指标 | 数值 | |----------------------|---------------| | **正确样本数** | 549 | | **错误样本数** | 448 | | **异常数** | 3 | | **总样本量** | 1000 | | **Python工具调用次数**| 889 | | **总补全Token数** | 1,301,333 | 原始统计数据: json { "accuracy": 0.549, "correct": 549, "incorrect": 448, "error": 3, "total": 1000, "python_tool_calls": 889, "completion_tokens": 1301333 }




