open-llm-leaderboard-old/details_SF-Foundation__TextBase-v0.2
收藏数据集概述
数据集基本信息
- 名称: Evaluation run of SF-Foundation/TextBase-v0.2
- 来源: 自动创建于模型 SF-Foundation/TextBase-v0.2 在 Open LLM Leaderboard 的评估运行中。
- 组成: 包含 63 个配置,每个配置对应一个评估任务。
- 创建次数: 从 2 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,用于计算和显示在 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_SF-Foundation__TextBase-v0.2", "harness_winogrande_5", split="train")
最新结果
以下是 2024-04-15T10:49:47.422583 运行的最新结果: python { "all": { "acc": 0.6477872383727521, "acc_stderr": 0.032222544304474854, "acc_norm": 0.6467665827432066, "acc_norm_stderr": 0.03290217364404172, "mc1": 0.6376988984088128, "mc1_stderr": 0.01682664689726226, "mc2": 0.7780333506353068, "mc2_stderr": 0.013795197050693505 }, "harness|arc:challenge|25": { "acc": 0.7244027303754266, "acc_stderr": 0.013057169655761838, "acc_norm": 0.7372013651877133, "acc_norm_stderr": 0.012862523175351333 }, "harness|hellaswag|10": { "acc": 0.7206731726747659, "acc_stderr": 0.004477514681328156, "acc_norm": 0.8897629954192392, "acc_norm_stderr": 0.0031254487960063553 }, "harness|hendrycksTest-abstract_algebra|5": { "acc": 0.35, "acc_stderr": 0.047937248544110196, "acc_norm": 0.35, "acc_norm_stderr": 0.047937248544110196 }, # 其他任务的结果... }
配置信息
-
配置名称: harness_arc_challenge_25
- 分割: 2024_04_15T10_42_26.389102
- 路径:
**/details_harness|arc:challenge|25_2024-04-15T10-42-26.389102.parquet
- 路径:
- 分割: 2024_04_15T10_49_47.422583
- 路径:
**/details_harness|arc:challenge|25_2024-04-15T10-49-47.422583.parquet
- 路径:
- 分割: latest
- 路径:
**/details_harness|arc:challenge|25_2024-04-15T10-49-47.422583.parquet
- 路径:
- 分割: 2024_04_15T10_42_26.389102
-
配置名称: harness_gsm8k_5
- 分割: 2024_04_15T10_42_26.389102
- 路径:
**/details_harness|gsm8k|5_2024-04-15T10-42-26.389102.parquet
- 路径:
- 分割: 2024_04_15T10_49_47.422583
- 路径:
**/details_harness|gsm8k|5_2024-04-15T10-49-47.422583.parquet
- 路径:
- 分割: latest
- 路径:
**/details_harness|gsm8k|5_2024-04-15T10-49-47.422583.parquet
- 路径:
- 分割: 2024_04_15T10_42_26.389102
-
配置名称: harness_hellaswag_10
- 分割: 2024_04_15T10_42_26.389102
- 路径:
**/details_harness|hellaswag|10_2024-04-15T10-42-26.389102.parquet
- 路径:
- 分割: 2024_04_15T10_49_47.422583
- 路径:
**/details_harness|hellaswag|10_2024-04-15T10-49-47.422583.parquet
- 路径:
- 分割: latest
- 路径:
**/details_harness|hellaswag|10_2024-04-15T10-49-47.422583.parquet
- 路径:
- 分割: 2024_04_15T10_42_26.389102
-
配置名称: harness_hendrycksTest_5
- 分割: 2024_04_15T10_42_26.389102
- 路径:
**/details_harness|hendrycksTest-abstract_algebra|5_2024-04-15T10-42-26.389102.parquet**/details_harness|hendrycksTest-anatomy|5_2024-04-15T10-42-26.389102.parquet**/details_harness|hendrycksTest-astronomy|5_2024-04-15T10-42-26.389102.parquet**/details_harness|hendrycksTest-business_ethics|5_2024-04-15T10-42-26.389102.parquet**/details_harness|hendrycksTest-clinical_knowledge|5_2024-04-15T10-42-26.389102.parquet**/details_harness|hendrycksTest-college_biology|5_2024-04-15T10-42-26.389102.parquet**/details_harness|hendrycksTest-college_chemistry|5_2024-04-15T10-42-26.389102.parquet**/details_harness|hendrycksTest-college_computer_science|5_2024-04-15T10-42-26.389102.parquet**/details_harness|hendrycksTest-college_mathematics|5_2024-04-15T10-42-26.389102.parquet**/details_harness|hendrycksTest-college_medicine|5_2024-04-15T10-42-26.389102.parquet**/details_harness|hendrycksTest-college_physics|5_2024-04-15T10-42-26.389102.parquet**/details_harness|hendrycksTest-computer_security|5_2024-04-15T10-42-26.389102.parquet**/details_harness|hendrycksTest-conceptual_physics|5_2024-04-15T10-42-26.389102.parquet**/details_harness|hendrycksTest-econometrics|5_2024-04-15T10-42-26.389102.parquet**/details_harness|hendrycksTest-electrical_engineering|5_2024-04-15T10-42-26.389102.parquet
- 路径:
- 分割: 2024_04_15T10_42_26.389102



