open-llm-leaderboard-old/details_CHIH-HUNG__llama-2-13b-FINETUNE3_3.3w-r4-gate_up_down
收藏数据集概述
该数据集是在评估模型 CHIH-HUNG/llama-2-13b-FINETUNE3_3.3w-r4-gate_up_down 在 Open LLM Leaderboard 上的自动创建的。
数据集组成
- 数据集包含 64 个配置,每个配置对应一个评估任务。
- 数据集从 2 次运行中创建,每个运行可以在每个配置中作为一个特定的分片找到,分片名称使用运行的时间戳。
- "train" 分片始终指向最新的结果。
额外配置
- 一个额外的配置 "results" 存储所有运行的聚合结果,用于计算和显示在 Open LLM Leaderboard 上的聚合指标。
加载数据示例
python from datasets import load_dataset data = load_dataset("open-llm-leaderboard/details_CHIH-HUNG__llama-2-13b-FINETUNE3_3.3w-r4-gate_up_down", "harness_winogrande_5", split="train")
最新结果
以下是 2023-10-29T06:58:27.684723 运行的最新结果: python { "all": { "em": 0.24486157718120805, "em_stderr": 0.004403654691385417, "f1": 0.2882906879194633, "f1_stderr": 0.004368960720592288, "acc": 0.44466102551920117, "acc_stderr": 0.010390042784194857 }, "harness|drop|3": { "em": 0.24486157718120805, "em_stderr": 0.004403654691385417, "f1": 0.2882906879194633, "f1_stderr": 0.004368960720592288 }, "harness|gsm8k|5": { "acc": 0.1197877179681577, "acc_stderr": 0.00894421340355304 }, "harness|winogrande|5": { "acc": 0.7695343330702447, "acc_stderr": 0.011835872164836673 } }
配置详情
-
harness_arc_challenge_25
- 分片:2023_10_01T14_29_00.192136
- 路径:
**/details_harness|arc:challenge|25_2023-10-01T14-29-00.192136.parquet - 分片:latest
- 路径:
**/details_harness|arc:challenge|25_2023-10-01T14-29-00.192136.parquet
-
harness_drop_3
- 分片:2023_10_29T06_58_27.684723
- 路径:
**/details_harness|drop|3_2023-10-29T06-58-27.684723.parquet - 分片:latest
- 路径:
**/details_harness|drop|3_2023-10-29T06-58-27.684723.parquet
-
harness_gsm8k_5
- 分片:2023_10_29T06_58_27.684723
- 路径:
**/details_harness|gsm8k|5_2023-10-29T06-58-27.684723.parquet - 分片:latest
- 路径:
**/details_harness|gsm8k|5_2023-10-29T06-58-27.684723.parquet
-
harness_hellaswag_10
- 分片:2023_10_01T14_29_00.192136
- 路径:
**/details_harness|hellaswag|10_2023-10-01T14-29-00.192136.parquet - 分片:latest
- 路径:
**/details_harness|hellaswag|10_2023-10-01T14-29-00.192136.parquet
-
harness_hendrycksTest_5
- 分片:2023_10_01T14_29_00.192136
- 路径:
**/details_harness|hendrycksTest-abstract_algebra|5_2023-10-01T14-29-00.192136.parquet等 38 个文件 - 分片:latest
- 路径:
**/details_harness|hendrycksTest-abstract_algebra|5_2023-10-01T14-29-00.192136.parquet等 38 个文件
-
harness_hendrycksTest_abstract_algebra_5
- 分片:2023_10_01T14_29_00.192136
- 路径:
**/details_harness|hendrycksTest-abstract_algebra|5_2023-10-01T14-29-00.192136.parquet - 分片:latest
- 路径:
**/details_harness|hendrycksTest-abstract_algebra|5_2023-10-01T14-29-00.192136.parquet
-
harness_hendrycksTest_anatomy_5
- 分片:2023_10_01T14_29_00.192136
- 路径:
**/details_harness|hendrycksTest-anatomy|5_2023-10-01T14-29-00.192136.parquet - 分片:latest
- 路径:
**/details_harness|hendrycksTest-anatomy|5_2023-10-01T14-29-00.192136.parquet
-
harness_hendrycksTest_astronomy_5
- 分片:2023_10_01T14_29_00.192136
- 路径:
**/details_harness|hendrycksTest-astronomy|5_2023-10-01T14-29-00.192136.parquet - 分片:latest
- 路径:
**/details_harness|hendrycksTest-astronomy|5_2023-10-01T14-29-00.192136.parquet
-
harness_hendrycksTest_business_ethics_5
- 分片:2023_10_01T14_29_00.192136
- 路径:
**/details_harness|hendrycksTest-business_ethics|5_2023-10-01T14-29-00.192136.parquet - 分片:latest
- 路径:
**/details_harness|hendrycksTest-business_ethics|5_2023-10-01T14-29-00.192136.parquet
-
harness_hendrycksTest_clinical_knowledge_5
- 分片:2023_10_01T14_29_00.192136
- 路径:
**/details_harness|hendrycksTest-clinical_knowledge|5_2023-10-01T14-29-00.192136.parquet - 分片:latest
- 路径:
**/details_harness|hendrycksTest-clinical_knowledge|5_2023-10-01T14-29-00.192136.parquet
-
harness_hendrycksTest_college_biology_5
- 分片:2023_10_01T14_29_00.192136
- 路径:
**/details_harness|hendrycksTest-college_biology|5_2023-10-01T14-29-00.192136.parquet - 分片:latest
- 路径:
**/details_harness|hendrycksTest-college_biology|5_2023-10-01T14-29-00.192136.parquet
-
harness_hendrycksTest_college_chemistry_5
- 分片:2023_10_01T14_29_00.192136
- 路径:
**/details_harness|hendrycksTest-college_chemistry|5_2023-10-01T14-29-00.192136.parquet - 分片:latest
- 路径:
**/details_harness|hendrycksTest-college_chemistry|5_2023-10-01T14-29-00.192136.parquet
-
harness_hendrycksTest_college_computer_science_5
- 分片:2023_10_01T14_29_00.192136
- 路径:
**/details_harness|hendrycksTest-college_computer_science|5_2023-10-01T14-29-00.192136.parquet - 分片:latest
- 路径:
**/details_harness|hendrycksTest-college_computer_science|5_2023-10-01T14-29-00.192136.parquet
-
harness_hendrycksTest_college_mathematics_5
- 分片:2023_10_01T14_29_00.192136
- 路径:
**/details_harness|hendrycksTest-college_mathematics|5_2023-10-01T14-29-00.192136.parquet - 分片:latest
- 路径:
**/details_harness|hendrycksTest-college_mathematics|5_2023-10-01T14-29-00.192136.parquet
-
harness_hendrycksTest_college_medicine_5
- 分片:2023_10_01T14_29_00.192136
- 路径:
**/details_harness|hendrycksTest-college_medicine|5_2023-10-01T14-29-00.192136.parquet - 分片:latest
- 路径:
**/details_harness|hendrycksTest-college_medicine|5_2023-10-01T14-29-00.192136.parquet
-
harness_hendrycksTest_college_physics_5
- 分片:2023_10_01T14_29_00.192136
- 路径:
**/details_harness|hendrycksTest-college_physics|5_2023-10-01T14-29-00.192136.parquet - 分片:latest
- 路径:
**/details_harness|hendrycksTest-college_physics|5_2023-10-01T14-29-00.192136.parquet
-
harness_hendrycksTest_computer_security_5
- 分片:2023_10_01T14_29_00.192136
- 路径:
**/details_harness|hendrycksTest-computer_security|5_2023-10-01T14-29-00.192136.parquet - 分片:latest
- 路径:
**/details_harness|hendrycksTest-computer_security|5_2023-10-01T14-29-00.192136.parquet
-
harness_hendrycksTest_conceptual_physics_5
- 分片:2023_10_01T14_29_00.192136
- 路径:
**/details_harness|hendrycksTest-conceptual_physics|5_2023-10-01T14-29-00.192136.parquet - 分片:latest
- 路径:
**/details_harness|hendrycksTest-conceptual_physics|5_2023-10-01T14-29-00.192136.parquet
-
harness_hendrycksTest_econometrics_5
- 分片:2023_10_01T14_29_00.192136
- 路径:
**/details_harness|hendrycksTest-econometrics|5_2023-10-01T14-29-00.192136.parquet - 分片:latest
- 路径:
**/details_harness|hendrycksTest-econometrics|5_2023-10-01T14-29-00.192136.parquet
-
**harness_hendrycksTest



