open-llm-leaderboard-old/details_openthaigpt__openthaigpt-1.0.0-alpha-7b-chat-ckpt-hf
收藏Hugging Face2023-09-22 更新2024-06-22 收录
下载链接:
https://hf-mirror.com/datasets/open-llm-leaderboard-old/details_openthaigpt__openthaigpt-1.0.0-alpha-7b-chat-ckpt-hf
下载链接
链接失效反馈官方服务:
资源简介:
该数据集是在模型 openthaigpt/openthaigpt-1.0.0-alpha-7b-chat-ckpt-hf 在 Open LLM Leaderboard 上的评估运行期间自动创建的。数据集由 64 个配置组成,每个配置对应一个被评估的任务。数据集是从 2 次运行中创建的,每次运行在每个配置中表示为特定的拆分。train 拆分始终指向最新的结果。一个名为 results 的额外配置存储了所有运行的聚合结果,这些结果用于计算和显示 Open LLM Leaderboard 上的聚合指标。README 还提供了一个示例,展示了如何使用 Python 中的 datasets 库加载运行中的详细信息。
该数据集是在模型 openthaigpt/openthaigpt-1.0.0-alpha-7b-chat-ckpt-hf 在 Open LLM Leaderboard 上的评估运行期间自动创建的。数据集由 64 个配置组成,每个配置对应一个被评估的任务。数据集是从 2 次运行中创建的,每次运行在每个配置中表示为特定的拆分。train 拆分始终指向最新的结果。一个名为 results 的额外配置存储了所有运行的聚合结果,这些结果用于计算和显示 Open LLM Leaderboard 上的聚合指标。README 还提供了一个示例,展示了如何使用 Python 中的 datasets 库加载运行中的详细信息。
提供机构:
open-llm-leaderboard-old
原始信息汇总
数据集概述
数据集名称
- Evaluation run of openthaigpt/openthaigpt-1.0.0-alpha-7b-chat-ckpt-hf
数据集来源
- 该数据集是在评估模型 openthaigpt/openthaigpt-1.0.0-alpha-7b-chat-ckpt-hf 在 Open LLM Leaderboard 上的自动创建的。
数据集结构
- 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" 分割总是指向最新的结果。
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。
数据加载示例
python from datasets import load_dataset data = load_dataset("open-llm-leaderboard/details_openthaigpt__openthaigpt-1.0.0-alpha-7b-chat-ckpt-hf", "harness_winogrande_5", split="train")
最新结果
- 这些是最新结果来自 run 2023-09-22T23:15:18.463104: python { "all": { "em": 0.031774328859060404, "em_stderr": 0.0017962473521312393, "f1": 0.08420092281879202, "f1_stderr": 0.0021474530604162255, "acc": 0.3646366953032391, "acc_stderr": 0.00915095624646051 }, "harness|drop|3": { "em": 0.031774328859060404, "em_stderr": 0.0017962473521312393, "f1": 0.08420092281879202, "f1_stderr": 0.0021474530604162255 }, "harness|gsm8k|5": { "acc": 0.03866565579984837, "acc_stderr": 0.005310583162098024 }, "harness|winogrande|5": { "acc": 0.6906077348066298, "acc_stderr": 0.012991329330822995 } }
配置详情
-
harness_arc_challenge_25
- 分割:2023_08_18T12_43_45.904593
- 路径:
**/details_harness|arc:challenge|25_2023-08-18T12:43:45.904593.parquet
- 路径:
- 分割:latest
- 路径:
**/details_harness|arc:challenge|25_2023-08-18T12:43:45.904593.parquet
- 路径:
- 分割:2023_08_18T12_43_45.904593
-
harness_drop_3
- 分割:2023_09_22T23_15_18.463104
- 路径:
**/details_harness|drop|3_2023-09-22T23-15-18.463104.parquet
- 路径:
- 分割:latest
- 路径:
**/details_harness|drop|3_2023-09-22T23-15-18.463104.parquet
- 路径:
- 分割:2023_09_22T23_15_18.463104
-
harness_gsm8k_5
- 分割:2023_09_22T23_15_18.463104
- 路径:
**/details_harness|gsm8k|5_2023-09-22T23-15-18.463104.parquet
- 路径:
- 分割:latest
- 路径:
**/details_harness|gsm8k|5_2023-09-22T23-15-18.463104.parquet
- 路径:
- 分割:2023_09_22T23_15_18.463104
-
harness_hellaswag_10
- 分割:2023_08_18T12_43_45.904593
- 路径:
**/details_harness|hellaswag|10_2023-08-18T12:43:45.904593.parquet
- 路径:
- 分割:latest
- 路径:
**/details_harness|hellaswag|10_2023-08-18T12:43:45.904593.parquet
- 路径:
- 分割:2023_08_18T12_43_45.904593
-
harness_hendrycksTest_5
- 分割:2023_08_18T12_43_45.904593
- 路径:
**/details_harness|hendrycksTest-abstract_algebra|5_2023-08-18T12:43:45.904593.parquet**/details_harness|hendrycksTest-anatomy|5_2023-08-18T12:43:45.904593.parquet**/details_harness|hendrycksTest-astronomy|5_2023-08-18T12:43:45.904593.parquet**/details_harness|hendrycksTest-business_ethics|5_2023-08-18T12:43:45.904593.parquet**/details_harness|hendrycksTest-clinical_knowledge|5_2023-08-18T12:43:45.904593.parquet**/details_harness|hendrycksTest-college_biology|5_2023-08-18T12:43:45.904593.parquet**/details_harness|hendrycksTest-college_chemistry|5_2023-08-18T12:43:45.904593.parquet**/details_harness|hendrycksTest-college_computer_science|5_2023-08-18T12:43:45.904593.parquet**/details_harness|hendrycksTest-college_mathematics|5_2023-08-18T12:43:45.904593.parquet**/details_harness|hendrycksTest-college_medicine|5_2023-08-18T12:43:45.904593.parquet**/details_harness|hendrycksTest-college_physics|5_2023-08-18T12:43:45.904593.parquet**/details_harness|hendrycksTest-computer_security|5_2023-08-18T12:43:45.904593.parquet**/details_harness|hendrycksTest-conceptual_physics|5_2023-08-18T12:43:45.904593.parquet**/details_harness|hendrycksTest-econometrics|5_2023-08-18T12:43:45.904593.parquet**/details_harness|hendrycksTest-electrical_engineering|5_2023-08-18T12:43:45.904593.parquet**/details_harness|hendrycksTest-elementary_mathematics|5_2023-08-18T12:43:45.904593.parquet**/details_harness|hendrycksTest-formal_logic|5_2023-08-18T12:43:45.904593.parquet**/details_harness|hendrycksTest-global_facts|5_2023-08-18T12:43:45.904593.parquet**/details_harness|hendrycksTest-high_school_biology|5_2023-08-18T12:43:45.904593.parquet**/details_harness|hendrycksTest-high_school_chemistry|5_2023-08-18T12:43:45.904593.parquet**/details_harness|hendrycksTest-high_school_computer_science|5_2023-08-18T12:43:45.904593.parquet**/details_harness|hendrycksTest-high_school_european_history|5_2023-08-18T12:43:45.904593.parquet**/details_harness|hendrycksTest-high_school_geography|5_2023-08-18T12:43:45.904593.parquet**/details_harness|hendrycksTest-high_school_government_and_politics|5_2023-08-18T12:43:45.904593.parquet**/details_harness|hendrycksTest-high_school_macroeconomics|5_2023-08-18T12:43:45.904593.parquet**/details_harness|hendrycksTest-high_school_mathematics|5_2023-08-18T12:43:45.904593.parquet**/details_harness|hendrycksTest-high_school_microeconomics|5_2023-08-18T12:43:45.904593.parquet**/details_harness|hendrycksTest-high_school_physics|5_2023-08-18T12:43:45.904593.parquet**/details_harness|hendrycksTest-high_school_psychology|5_2023-08-18T12:43:45.904593.parquet**/details_harness|hendrycksTest-high_school_statistics|5_2023-08-18T12:43:45.904593.parquet**/details_harness|hendrycksTest-high_school_us_history|5_2023-08-18T12:43:45.904593.parquet**/details_harness|hendrycksTest-high_school_world_history|5_2023-08-18T12:43:45.904593.parquet**/details_harness|hendrycksTest-human_aging|5_2023-08-18T12:43:45.904593.parquet**/details_harness|hendrycksTest-human_sexuality|5_2023-08-18T12:43:45.904593.parquet**/details_harness|hendrycksTest-international_law|5_2023-08-18T12:43:45.904593.parquet**/details_harness|hendrycksTest-jurisprudence|5_2023-08-18T12:43:45.904593.parquet**/details_harness|hendrycksTest-logical_fallacies|5_2023-08-18T12:43:45.904593.parquet**/details_harness|hendrycksTest-machine_learning|5_2023-08-18T12:43:45.904593.parquet**/details_harness|hendrycksTest-management|5_2023-08-18T12:43:45.904593.parquet**/details_harness|hendrycksTest-marketing|5_2023-08-18T12:43:45.904593.parquet**/details_harness|hendrycksTest-medical_genetics|5_2023-08-18T12:43:45.904593.parquet**/details_harness|hendrycksTest-miscellaneous|5_2023-08-18T12:43:45.904593.parquet**/details_harness|hendrycksTest-moral_disputes|5_2023-08-18T12:43:45.904593.parquet**/details_harness|hendrycksTest-moral_scenarios|5_2023-08-18T12:43:45.904593.parquet**/details_harness|hendrycksTest-nutrition|5_2023-08-18T12:43:45.904593.parquet**/details_harness|hendrycksTest-philosophy|5_2023-08-18T12:43:45.904593.parquet**/details_harness|hendrycksTest-prehistory|5_2023-08-18T12:43:45.904593.parquet- `**/details_
- 路径:
- 分割:2023_08_18T12_43_45.904593



