open-llm-leaderboard-old/details_FelixChao__llama2-13b-math1.2
收藏Hugging Face2023-10-17 更新2024-06-22 收录
下载链接：
https://hf-mirror.com/datasets/open-llm-leaderboard-old/details_FelixChao__llama2-13b-math1.2
下载链接
链接失效反馈官方服务：
资源简介：
该数据集是在评估模型FelixChao/llama2-13b-math1.2时自动生成的，评估过程在Open LLM Leaderboard上进行。数据集由64个配置组成，每个配置对应一个评估任务。数据集由3次运行生成，每次运行的结果作为特定配置中的一个分割，分割名称使用运行的时间戳。train分割始终指向最新的结果。此外，还有一个名为results的配置，存储了所有运行的聚合结果，并用于计算和显示Open LLM Leaderboard上的聚合指标。
该数据集是在评估模型FelixChao/llama2-13b-math1.2时自动生成的，评估过程在Open LLM Leaderboard上进行。数据集由64个配置组成，每个配置对应一个评估任务。数据集由3次运行生成，每次运行的结果作为特定配置中的一个分割，分割名称使用运行的时间戳。train分割始终指向最新的结果。此外，还有一个名为results的配置，存储了所有运行的聚合结果，并用于计算和显示Open LLM Leaderboard上的聚合指标。
提供机构：
open-llm-leaderboard-old
原始信息汇总
数据集概述
数据集简介
该数据集是在对模型 FelixChao/llama2-13b-math1.2 进行评估运行期间自动创建的，用于 Open LLM Leaderboard。
数据集结构
- 配置数量：64个配置，每个配置对应一个评估任务。
- 创建来源：从3次运行中创建。每个运行对应一个特定的分割，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，用于计算和显示在 Open LLM Leaderboard 上的汇总指标。
数据加载示例
python from datasets import load_dataset data = load_dataset("open-llm-leaderboard/details_FelixChao__llama2-13b-math1.2", "harness_winogrande_5", split="train")
最新结果
以下是 2023-10-17T02:02:50.506714 运行的最新结果： python { "all": { "em": 0.08536073825503356, "em_stderr": 0.002861499356149465, "f1": 0.16235318791946293, "f1_stderr": 0.003133455634092774, "acc": 0.4263155280751903, "acc_stderr": 0.010451092603365564 }, "harness|drop|3": { "em": 0.08536073825503356, "em_stderr": 0.002861499356149465, "f1": 0.16235318791946293, "f1_stderr": 0.003133455634092774 }, "harness|gsm8k|5": { "acc": 0.10993176648976498, "acc_stderr": 0.008616195587865416 }, "harness|winogrande|5": { "acc": 0.7426992896606156, "acc_stderr": 0.012285989618865713 } }
配置详情
以下是部分配置的详细信息：
配置 harness_arc_challenge_25
- 分割：2023_08_18T11_24_31.239858
- 路径：
**/details_harness|arc:challenge|25_2023-08-18T11:24:31.239858.parquet
- 路径：
- 分割：latest
- 路径：
**/details_harness|arc:challenge|25_2023-08-18T11:24:31.239858.parquet
- 路径：
配置 harness_drop_3
- 分割：2023_10_16T16_38_50.665467
- 路径：
**/details_harness|drop|3_2023-10-16T16-38-50.665467.parquet
- 路径：
- 分割：2023_10_17T02_02_50.506714
- 路径：
**/details_harness|drop|3_2023-10-17T02-02-50.506714.parquet
- 路径：
- 分割：latest
- 路径：
**/details_harness|drop|3_2023-10-17T02-02-50.506714.parquet
- 路径：
配置 harness_gsm8k_5
- 分割：2023_10_16T16_38_50.665467
- 路径：
**/details_harness|gsm8k|5_2023-10-16T16-38-50.665467.parquet
- 路径：
- 分割：2023_10_17T02_02_50.506714
- 路径：
**/details_harness|gsm8k|5_2023-10-17T02-02-50.506714.parquet
- 路径：
- 分割：latest
- 路径：
**/details_harness|gsm8k|5_2023-10-17T02-02-50.506714.parquet
- 路径：
配置 harness_hellaswag_10
- 分割：2023_08_18T11_24_31.239858
- 路径：
**/details_harness|hellaswag|10_2023-08-18T11:24:31.239858.parquet
- 路径：
- 分割：latest
- 路径：
**/details_harness|hellaswag|10_2023-08-18T11:24:31.239858.parquet
- 路径：
配置 harness_hendrycksTest_5
- 分割：2023_08_18T11_24_31.239858
- 路径：
**/details_harness|hendrycksTest-abstract_algebra|5_2023-08-18T11:24:31.239858.parquet**/details_harness|hendrycksTest-anatomy|5_2023-08-18T11:24:31.239858.parquet**/details_harness|hendrycksTest-astronomy|5_2023-08-18T11:24:31.239858.parquet**/details_harness|hendrycksTest-business_ethics|5_2023-08-18T11:24:31.239858.parquet**/details_harness|hendrycksTest-clinical_knowledge|5_2023-08-18T11:24:31.239858.parquet**/details_harness|hendrycksTest-college_biology|5_2023-08-18T11:24:31.239858.parquet**/details_harness|hendrycksTest-college_chemistry|5_2023-08-18T11:24:31.239858.parquet**/details_harness|hendrycksTest-college_computer_science|5_2023-08-18T11:24:31.239858.parquet**/details_harness|hendrycksTest-college_mathematics|5_2023-08-18T11:24:31.239858.parquet**/details_harness|hendrycksTest-college_medicine|5_2023-08-18T11:24:31.239858.parquet**/details_harness|hendrycksTest-college_physics|5_2023-08-18T11:24:31.239858.parquet**/details_harness|hendrycksTest-computer_security|5_2023-08-18T11:24:31.239858.parquet**/details_harness|hendrycksTest-conceptual_physics|5_2023-08-18T11:24:31.239858.parquet**/details_harness|hendrycksTest-econometrics|5_2023-08-18T11:24:31.239858.parquet**/details_harness|hendrycksTest-electrical_engineering|5_2023-08-18T11:24:31.239858.parquet**/details_harness|hendrycksTest-elementary_mathematics|5_2023-08-18T11:24:31.239858.parquet**/details_harness|hendrycksTest-formal_logic|5_2023-08-18T11:24:31.239858.parquet**/details_harness|hendrycksTest-global_facts|5_2023-08-18T11:24:31.239858.parquet**/details_harness|hendrycksTest-high_school_biology|5_2023-08-18T11:24:31.239858.parquet**/details_harness|hendrycksTest-high_school_chemistry|5_2023-08-18T11:24:31.239858.parquet**/details_harness|hendrycksTest-high_school_computer_science|5_2023-08-18T11:24:31.239858.parquet**/details_harness|hendrycksTest-high_school_european_history|5_2023-08-18T11:24:31.239858.parquet**/details_harness|hendrycksTest-high_school_geography|5_2023-08-18T11:24:31.239858.parquet**/details_harness|hendrycksTest-high_school_government_and_politics|5_2023-08-18T11:24:31.239858.parquet**/details_harness|hendrycksTest-high_school_macroeconomics|5_2023-08-18T11:24:31.239858.parquet**/details_harness|hendrycksTest-high_school_mathematics|5_2023-08-18T11:24:31.239858.parquet**/details_harness|hendrycksTest-high_school_microeconomics|5_2023-08-18T11:24:31.239858.parquet**/details_harness|hendrycksTest-high_school_physics|5_2023-08-18T11:24:31.239858.parquet**/details_harness|hendrycksTest-high_school_psychology|5_2023-08-18T11:24:31.239858.parquet**/details_harness|hendrycksTest-high_school_statistics|5_2023-08-18T11:24:31.239858.parquet**/details_harness|hendrycksTest-high_school_us_history|5_2023-08-18T11:24:31.239858.parquet**/details_harness|hendrycksTest-high_school_world_history|5_2023-08-18T11:24:31.239858.parquet**/details_harness|hendrycksTest-human_aging|5_2023-08-18T11:24:31.239858.parquet**/details_harness|hendrycksTest-human_sexuality|5_2023-08-18T11:24:31.239858.parquet**/details_harness|hendrycksTest-international_law|5_2023-08-18T11:24:31.239858.parquet**/details_harness|hendrycksTest-jurisprudence|5_2023-08-18T11:24:31.239858.parquet**/details_harness|hendrycksTest-logical_fallacies|5_2023-08-18T11:24:31.239858.parquet**/details_harness|hendrycksTest-machine_learning|5_2023-08-18T11:24:31.239858.parquet**/details_harness|hendrycksTest-management|5_2023-08-18T11:24:31.239858.parquet**/details_harness|hendrycksTest-marketing|5_2023-08-18T11:24:31.239858.parquet**/details_harness|hendrycksTest-medical_genetics|5_2023-08-18T11:24:31.239858.parquet**/details_harness|hendrycksTest-miscellaneous|5_2023-08-18T11:24:31.239858.parquet**/details_harness|hendrycksTest-moral_disputes|5_2023-08-18T11:24:31.239858.parquet**/details_harness|hendrycksTest-moral_scenarios|5_2023-08-18T11:24:31.239858.parquet- `**/details_harness|hendrycksTest-nutrition|5_2023-08-18T
- 路径：
