open-llm-leaderboard-old/details_JCX-kcuf__Llama-2-7b-hf-gpt-4-80k
收藏数据集概述
数据集摘要
该数据集是在模型 JCX-kcuf/Llama-2-7b-hf-gpt-4-80k 在 Open LLM Leaderboard 上的评估运行期间自动创建的。
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。
- 创建来源:从1次运行中创建。每个运行可以在每个配置中找到一个特定的分割,分割名称使用运行的时间戳。
- 训练分割:"train" 分割始终指向最新结果。
- 结果配置:一个额外的配置 "results" 存储所有运行的聚合结果,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_JCX-kcuf__Llama-2-7b-hf-gpt-4-80k", "harness_winogrande_5", split="train")
最新结果
这些是最新结果,来自 2024-03-11T18:19:30.657181 运行:
python { "all": { "acc": 0.4691427623076365, "acc_stderr": 0.03458340541870251, "acc_norm": 0.4741335558049342, "acc_norm_stderr": 0.035370693026626564, "mc1": 0.3390452876376989, "mc1_stderr": 0.016571797910626615, "mc2": 0.4862513962984137, "mc2_stderr": 0.014967511131756355 }, "harness|arc:challenge|25": { "acc": 0.5110921501706485, "acc_stderr": 0.01460779491401305, "acc_norm": 0.5554607508532423, "acc_norm_stderr": 0.01452122640562708 }, "harness|hellaswag|10": { "acc": 0.5802628958374826, "acc_stderr": 0.0049250721597238365, "acc_norm": 0.7726548496315475, "acc_norm_stderr": 0.004182607685205692 }, "harness|hendrycksTest-abstract_algebra|5": { "acc": 0.3, "acc_stderr": 0.046056618647183814, "acc_norm": 0.3, "acc_norm_stderr": 0.046056618647183814 }, ... }
配置详情
-
config_name: harness_arc_challenge_25
- data_files:
- split: 2024_03_11T18_19_30.657181
- **/details_harness|arc:challenge|25_2024-03-11T18-19-30.657181.parquet
- split: latest
- **/details_harness|arc:challenge|25_2024-03-11T18-19-30.657181.parquet
- split: 2024_03_11T18_19_30.657181
- data_files:
-
config_name: harness_gsm8k_5
- data_files:
- split: 2024_03_11T18_19_30.657181
- **/details_harness|gsm8k|5_2024-03-11T18-19-30.657181.parquet
- split: latest
- **/details_harness|gsm8k|5_2024-03-11T18-19-30.657181.parquet
- split: 2024_03_11T18_19_30.657181
- data_files:
-
config_name: harness_hellaswag_10
- data_files:
- split: 2024_03_11T18_19_30.657181
- **/details_harness|hellaswag|10_2024-03-11T18-19-30.657181.parquet
- split: latest
- **/details_harness|hellaswag|10_2024-03-11T18-19-30.657181.parquet
- split: 2024_03_11T18_19_30.657181
- data_files:
-
config_name: harness_hendrycksTest_5
- data_files:
- split: 2024_03_11T18_19_30.657181
- **/details_harness|hendrycksTest-abstract_algebra|5_2024-03-11T18-19-30.657181.parquet
- **/details_harness|hendrycksTest-anatomy|5_2024-03-11T18-19-30.657181.parquet
- **/details_harness|hendrycksTest-astronomy|5_2024-03-11T18-19-30.657181.parquet
- **/details_harness|hendrycksTest-business_ethics|5_2024-03-11T18-19-30.657181.parquet
- **/details_harness|hendrycksTest-clinical_knowledge|5_2024-03-11T18-19-30.657181.parquet
- **/details_harness|hendrycksTest-college_biology|5_2024-03-11T18-19-30.657181.parquet
- **/details_harness|hendrycksTest-college_chemistry|5_2024-03-11T18-19-30.657181.parquet
- **/details_harness|hendrycksTest-college_computer_science|5_2024-03-11T18-19-30.657181.parquet
- **/details_harness|hendrycksTest-college_mathematics|5_2024-03-11T18-19-30.657181.parquet
- **/details_harness|hendrycksTest-college_medicine|5_2024-03-11T18-19-30.657181.parquet
- **/details_harness|hendrycksTest-college_physics|5_2024-03-11T18-19-30.657181.parquet
- **/details_harness|hendrycksTest-computer_security|5_2024-03-11T18-19-30.657181.parquet
- **/details_harness|hendrycksTest-conceptual_physics|5_2024-03-11T18-19-30.657181.parquet
- **/details_harness|hendrycksTest-econometrics|5_2024-03-11T18-19-30.657181.parquet
- **/details_harness|hendrycksTest-electrical_engineering|5_2024-03-11T18-19-30.657181.parquet
- **/details_harness|hendrycksTest-elementary_mathematics|5_2024-03-11T18-19-30.657181.parquet
- **/details_harness|hendrycksTest-formal_logic|5_2024-03-11T18-19-30.657181.parquet
- **/details_harness|hendrycksTest-global_facts|5_2024-03-11T18-19-30.657181.parquet
- ...
- split: latest
- **/details_harness|hendrycksTest-abstract_algebra|5_2024-03-11T18-19-30.657181.parquet
- **/details_harness|hendrycksTest-anatomy|5_2024-03-11T18-19-30.657181.parquet
- **/details_harness|hendrycksTest-astronomy|5_2024-03-11T18-19-30.657181.parquet
- **/details_harness|hendrycksTest-business_ethics|5_2024-03-11T18-19-30.657181.parquet
- **/details_harness|hendrycksTest-clinical_knowledge|5_2024-03-11T18-19-30.657181.parquet
- **/details_harness|hendrycksTest-college_biology|5_2024-03-11T18-19-30.657181.parquet
- **/details_harness|hendrycksTest-college_chemistry|5_2024-03-11T18-19-30.657181.parquet
- **/details_harness|hendrycksTest-college_computer_science|5_2024-03-11T18-19-30.657181.parquet
- **/details_harness|hendrycksTest-college_mathematics|5_2024-03-11T18-19-30.657181.parquet
- **/details_harness|hendrycksTest-college_medicine|5_2024-03-11T18-19-30.657181.parquet
- **/details_harness|hendrycksTest-college_physics|5_2024-03-11T18-19-30.657181.parquet
- **/details_harness|hendrycksTest-computer_security|5_2024-03-11T18-19-30.657181.parquet
- **/details_harness|hendrycksTest-conceptual_physics|5_2024-03-11T18-19-30.657181.parquet
- **/details_harness|hendrycksTest-econometrics|5_2024-03-11T18-19-30.657181.parquet
- **/details_harness|hendrycksTest-electrical_engineering|5_2024-03-11T18-19-30.657181.parquet
- **/details_harness|hendrycksTest-elementary_mathematics|5_2024-03-11T18-19-30.657181.parquet
- **/details_harness|hendrycksTest-formal_logic|5_2024-03-11T18-19-30.657181.parquet
- **/details_harness|hendrycksTest-global_facts|5_2024-03-11T18-19-30.657181.parquet
- ...
- split: 2024_03_11T18_19_30.657181
- data_files:



