open-llm-leaderboard/details_togethercomputer__RedPajama-INCITE-Instruct-7B-v0.1

该数据集是在模型 togethercomputer/RedPajama-INCITE-Instruct-7B-v0.1 在 Open LLM Leaderboard 上的评估过程中自动生成的。数据集包含 64 个配置，每个配置对应一个评估任务。数据集由 2 次运行的数据组成，每次运行的数据作为每个配置中的一个特定分割，分割名称由运行的时间戳命名。train 分割始终指向最新的结果。此外，名为 results 的配置存储了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。README 文件还提供了一个 Python 代码片段来加载数据集详细信息，并列出了特定运行的最新结果。

This dataset was automatically generated during the evaluation of the model togethercomputer/RedPajama-INCITE-Instruct-7B-v0.1 on the Open LLM Leaderboard. It contains 64 configurations, each corresponding to one evaluation task. The dataset consists of data from two runs, where the data from each run acts as a specific split within each configuration, and the split names are named after the timestamp of the corresponding run. The "train" split always points to the most recent results. Additionally, the configuration named "results" stores the aggregated results across all runs, which are used to calculate and display the aggregate metrics on the Open LLM Leaderboard. The README file also provides a Python code snippet for loading detailed dataset information, and lists the latest results for specific runs.

原始信息汇总

数据集概述

该数据集是在对模型 togethercomputer/RedPajama-INCITE-Instruct-7B-v0.1 进行评估运行期间自动创建的，用于 Open LLM Leaderboard 上的评估。

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，每次运行可以在每个配置中找到特定的分割，分割名称使用运行的时间戳。
• "train" 分割始终指向最新的结果。
• 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_togethercomputer__RedPajama-INCITE-Instruct-7B-v0.1", "harness_winogrande_5", split="train")

最新结果

以下是 最新结果 的示例： python { "all": { "em": 0.001153523489932886, "em_stderr": 0.0003476179896857104, "f1": 0.04208578020134259, "f1_stderr": 0.00114625984545935, "acc": 0.3327435280488615, "acc_stderr": 0.008428433474529594 }, "harness|drop|3": { "em": 0.001153523489932886, "em_stderr": 0.0003476179896857104, "f1": 0.04208578020134259, "f1_stderr": 0.00114625984545935 }, "harness|gsm8k|5": { "acc": 0.01592115238817286, "acc_stderr": 0.0034478192723889985 }, "harness|winogrande|5": { "acc": 0.6495659037095501, "acc_stderr": 0.013409047676670187 } }

配置详情

• harness_arc_challenge_25

  • 分割: 2023_07_19T17_13_41.605696, latest
  • 路径: **/details_harness|arc:challenge|25_2023-07-19T17:13:41.605696.parquet

• harness_drop_3

  • 分割: 2023_10_21T19_03_28.692424, latest
  • 路径: **/details_harness|drop|3_2023-10-21T19-03-28.692424.parquet

• harness_gsm8k_5

  • 分割: 2023_10_21T19_03_28.692424, latest
  • 路径: **/details_harness|gsm8k|5_2023-10-21T19-03-28.692424.parquet

• harness_hellaswag_10

  • 分割: 2023_07_19T17_13_41.605696, latest
  • 路径: **/details_harness|hellaswag|10_2023-07-19T17:13:41.605696.parquet

• harness_hendrycksTest_5

  • 分割: 2023_07_19T17_13_41.605696, latest
  • 路径: 多个路径，例如 **/details_harness|hendrycksTest-abstract_algebra|5_2023-07-19T17:13:41.605696.parquet

• harness_hendrycksTest_abstract_algebra_5

  • 分割: 2023_07_19T17_13_41.605696, latest
  • 路径: **/details_harness|hendrycksTest-abstract_algebra|5_2023-07-19T17:13:41.605696.parquet

• harness_hendrycksTest_anatomy_5

  • 分割: 2023_07_19T17_13_41.605696, latest
  • 路径: **/details_harness|hendrycksTest-anatomy|5_2023-07-19T17:13:41.605696.parquet

• harness_hendrycksTest_astronomy_5

  • 分割: 2023_07_19T17_13_41.605696, latest
  • 路径: **/details_harness|hendrycksTest-astronomy|5_2023-07-19T17:13:41.605696.parquet

• harness_hendrycksTest_business_ethics_5

  • 分割: 2023_07_19T17_13_41.605696, latest
  • 路径: **/details_harness|hendrycksTest-business_ethics|5_2023-07-19T17:13:41.605696.parquet

• harness_hendrycksTest_clinical_knowledge_5

  • 分割: 2023_07_19T17_13_41.605696, latest
  • 路径: **/details_harness|hendrycksTest-clinical_knowledge|5_2023-07-19T17:13:41.605696.parquet

• harness_hendrycksTest_college_biology_5

  • 分割: 2023_07_19T17_13_41.605696, latest
  • 路径: **/details_harness|hendrycksTest-college_biology|5_2023-07-19T17:13:41.605696.parquet

• harness_hendrycksTest_college_chemistry_5

  • 分割: 2023_07_19T17_13_41.605696, latest
  • 路径: **/details_harness|hendrycksTest-college_chemistry|5_2023-07-19T17:13:41.605696.parquet

• harness_hendrycksTest_college_computer_science_5

  • 分割: 2023_07_19T17_13_41.605696, latest
  • 路径: **/details_harness|hendrycksTest-college_computer_science|5_2023-07-19T17:13:41.605696.parquet

• harness_hendrycksTest_college_mathematics_5

  • 分割: 2023_07_19T17_13_41.605696, latest
  • 路径: **/details_harness|hendrycksTest-college_mathematics|5_2023-07-19T17:13:41.605696.parquet

• harness_hendrycksTest_college_medicine_5

  • 分割: 2023_07_19T17_13_41.605696, latest
  • 路径: **/details_harness|hendrycksTest-college_medicine|5_2023-07-19T17:13:41.605696.parquet

• harness_hendrycksTest_college_physics_5

  • 分割: 2023_07_19T17_13_41.605696, latest
  • 路径: **/details_harness|hendrycksTest-college_physics|5_2023-07-19T17:13:41.605696.parquet

• harness_hendrycksTest_computer_security_5

  • 分割: 2023_07_19T17_13_41.605696, latest
  • 路径: **/details_harness|hendrycksTest-computer_security|5_2023-07-19T17:13:41.605696.parquet

• harness_hendrycksTest_conceptual_physics_5

  • 分割: 2023_07_19T17_13_41.605696, latest
  • 路径: **/details_harness|hendrycksTest-conceptual_physics|5_2023-07-19T17:13:41.605696.parquet

• harness_hendrycksTest_econometrics_5

  • 分割: 2023_07_19T17_13_41.605696, latest
  • 路径: **/details_harness|hendrycksTest-econometrics|5_2023-07-19T17:13:41.605696.parquet

• harness_hendrycksTest_electrical_engineering_5

  • 分割: 2023_07_19T17_13_41.605696, latest
  • 路径: **/details_harness|hendrycksTest-electrical_engineering|5_2023-07-19T17:13:41.605696.parquet