TrafficBenchV1/TrafficBench_Nov6th_v1.1
收藏Hugging Face2025-11-07 更新2025-11-15 收录
下载链接：
https://hf-mirror.com/datasets/TrafficBenchV1/TrafficBench_Nov6th_v1.1
下载链接
链接失效反馈官方服务：
资源简介：
该数据集包含了四个部分：wildchat、natural_reasoning、mmlu_pro和supergpqa，每个部分都包含10000个示例。数据集的字段包括问题、答案、主题以及关于数据集和模型性能的详细元数据。这些性能指标包括计算力、能耗、输入 tokens 数、延迟、内存带宽、GPU 数量和推理时间等。
The dataset consists of four parts: wildchat, natural_reasoning, mmlu_pro, and supergpqa, each containing 10,000 examples. The dataset fields include problem, answer, subject, and detailed metadata about the dataset and model performance. These performance metrics include compute power, energy consumption, input tokens count, latency, memory bandwidth, number of GPUs, and inference time.
提供机构：
TrafficBenchV1
