open-llm-leaderboard-old/details_mwitiderrick__shearedplats-2.7b-v2-instruct-v0.1
收藏数据集概述
数据集简介
该数据集是在评估模型 mwitiderrick/shearedplats-2.7b-v2-instruct-v0.1 在 Open LLM Leaderboard 上的运行过程中自动创建的。
数据集结构
- 配置数量:63个配置,每个配置对应一个评估任务。
- 数据来源:数据集从1次运行中创建,每个运行在每个配置中作为一个特定的分割存在,分割名称使用运行的时间戳。
- 最新结果:"train" 分割始终指向最新的结果。
- 结果汇总:一个额外的配置 "results" 存储所有运行的汇总结果,用于计算和显示 Open LLM Leaderboard 上的聚合指标。
数据加载示例
python from datasets import load_dataset data = load_dataset("open-llm-leaderboard/details_mwitiderrick__shearedplats-2.7b-v2-instruct-v0.1", "harness_winogrande_5", split="train")
最新结果
以下是 2023-12-09T18:49:52.400292 运行的最新结果:
python { "all": { "acc": 0.2882361931913223, "acc_stderr": 0.031895486998552665, "acc_norm": 0.2903928342274915, "acc_norm_stderr": 0.03267939625046512, "mc1": 0.2729498164014688, "mc1_stderr": 0.015594753632006535, "mc2": 0.41227748774876055, "mc2_stderr": 0.014572961912704371 }, "harness|arc:challenge|25": { "acc": 0.3660409556313993, "acc_stderr": 0.01407722310847014, "acc_norm": 0.40187713310580203, "acc_norm_stderr": 0.014327268614578274 }, "harness|hellaswag|10": { "acc": 0.5142401911969727, "acc_stderr": 0.00498775731476984, "acc_norm": 0.7007568213503286, "acc_norm_stderr": 0.00456990648509029 }, "harness|hendrycksTest-abstract_algebra|5": { "acc": 0.23, "acc_stderr": 0.04229525846816506, "acc_norm": 0.23, "acc_norm_stderr": 0.04229525846816506 }, # 其他任务的结果... }
配置详情
-
harness_arc_challenge_25
- 分割:2023_12_09T18_49_52.400292
- 路径:
**/details_harness|arc:challenge|25_2023-12-09T18-49-52.400292.parquet - 分割:latest
- 路径:
**/details_harness|arc:challenge|25_2023-12-09T18-49-52.400292.parquet
-
harness_gsm8k_5
- 分割:2023_12_09T18_49_52.400292
- 路径:
**/details_harness|gsm8k|5_2023-12-09T18-49-52.400292.parquet - 分割:latest
- 路径:
**/details_harness|gsm8k|5_2023-12-09T18-49-52.400292.parquet
-
harness_hellaswag_10
- 分割:2023_12_09T18_49_52.400292
- 路径:
**/details_harness|hellaswag|10_2023-12-09T18-49-52.400292.parquet - 分割:latest
- 路径:
**/details_harness|hellaswag|10_2023-12-09T18-49-52.400292.parquet
-
harness_hendrycksTest_5
- 分割:2023_12_09T18_49_52.400292
- 路径:
**/details_harness|hendrycksTest-abstract_algebra|5_2023-12-09T18-49-52.400292.parquet**/details_harness|hendrycksTest-anatomy|5_2023-12-09T18-49-52.400292.parquet**/details_harness|hendrycksTest-astronomy|5_2023-12-09T18-49-52.400292.parquet**/details_harness|hendrycksTest-business_ethics|5_2023-12-09T18-49-52.400292.parquet**/details_harness|hendrycksTest-clinical_knowledge|5_2023-12-09T18-49-52.400292.parquet**/details_harness|hendrycksTest-college_biology|5_2023-12-09T18-49-52.400292.parquet**/details_harness|hendrycksTest-college_chemistry|5_2023-12-09T18-49-52.400292.parquet**/details_harness|hendrycksTest-college_computer_science|5_2023-12-09T18-49-52.400292.parquet**/details_harness|hendrycksTest-college_mathematics|5_2023-12-09T18-49-52.400292.parquet**/details_harness|hendrycksTest-college_medicine|5_2023-12-09T18-49-52.400292.parquet**/details_harness|hendrycksTest-college_physics|5_2023-12-09T18-49-52.400292.parquet**/details_harness|hendrycksTest-computer_security|5_2023-12-09T18-49-52.400292.parquet**/details_harness|hendrycksTest-conceptual_physics|5_2023-12-09T18-49-52.400292.parquet**/details_harness|hendrycksTest-econometrics|5_2023-12-09T18-49-52.400292.parquet**/details_harness|hendrycksTest-electrical_engineering|5_2023-12-09T18-49-52.400292.parquet**/details_harness|hendrycksTest-elementary_mathematics|5_2023-12-09T18-49-52.400292.parquet**/details_harness|hendrycksTest-formal_logic|5_2023-12-09T18-49-52.400292.parquet
其他路径...
}



