open-llm-leaderboard-old/details_ibm__merlinite-7b
收藏Hugging Face2024-03-05 更新2024-06-22 收录
下载链接:
https://hf-mirror.com/datasets/open-llm-leaderboard-old/details_ibm__merlinite-7b
下载链接
链接失效反馈官方服务:
资源简介:
该数据集是在Open LLM Leaderboard上评估模型ibm/merlinite-7b时自动创建的。数据集包含63个配置,每个配置对应一个评估任务。数据集由1次运行生成,每次运行可以在每个配置中找到,运行的时间戳作为分割名称。train分割始终指向最新结果。results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。
该数据集是在Open LLM Leaderboard上评估模型ibm/merlinite-7b时自动创建的。数据集包含63个配置,每个配置对应一个评估任务。数据集由1次运行生成,每次运行可以在每个配置中找到,运行的时间戳作为分割名称。train分割始终指向最新结果。results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。
提供机构:
open-llm-leaderboard-old
原始信息汇总
数据集概述
数据集信息
- 名称: Evaluation run of ibm/merlinite-7b
- 概述: 该数据集是在模型 ibm/merlinite-7b 在 Open LLM Leaderboard 上的评估运行期间自动创建的。
- 组成: 包含 63 个配置,每个配置对应一个评估任务。
- 创建: 数据集从 1 次运行中创建,每个运行在每个配置中作为一个特定的分割存在,分割名称使用运行的时间戳。"train" 分割始终指向最新的结果。
- 结果配置: 一个额外的配置 "results" 存储所有运行的聚合结果,用于计算并在 Open LLM Leaderboard 上显示聚合指标。
数据加载示例
python from datasets import load_dataset data = load_dataset("open-llm-leaderboard/details_ibm__merlinite-7b", "harness_winogrande_5", split="train")
最新结果
- 最新结果时间戳: 2024-03-05T00:51:32.101998
- 结果示例: python { "all": { "acc": 0.6466281374367928, "acc_stderr": 0.03209827706115288, "acc_norm": 0.6521715694423292, "acc_norm_stderr": 0.03274036323874206, "mc1": 0.33659730722154224, "mc1_stderr": 0.016542412809494887, "mc2": 0.5014708864565086, "mc2_stderr": 0.014843262999616028 }, "harness|arc:challenge|25": { "acc": 0.5930034129692833, "acc_stderr": 0.014356399418009121, "acc_norm": 0.636518771331058, "acc_norm_stderr": 0.014056207319068283 }, "harness|hellaswag|10": { "acc": 0.6467835092611034, "acc_stderr": 0.004769924131304649, "acc_norm": 0.8451503684524995, "acc_norm_stderr": 0.003610219413061462 }, "harness|hendrycksTest-abstract_algebra|5": { "acc": 0.38, "acc_stderr": 0.04878317312145632, "acc_norm": 0.38, "acc_norm_stderr": 0.04878317312145632 }, # 其他任务的结果... }
配置详情
-
配置名称: harness_arc_challenge_25
- 分割: 2024_03_05T00_51_32.101998, latest
- 路径:
- **/details_harness|arc:challenge|25_2024-03-05T00-51-32.101998.parquet
-
配置名称: harness_gsm8k_5
- 分割: 2024_03_05T00_51_32.101998, latest
- 路径:
- **/details_harness|gsm8k|5_2024-03-05T00-51-32.101998.parquet
-
配置名称: harness_hellaswag_10
- 分割: 2024_03_05T00_51_32.101998, latest
- 路径:
- **/details_harness|hellaswag|10_2024-03-05T00-51-32.101998.parquet
-
配置名称: harness_hendrycksTest_5
- 分割: 2024_03_05T00_51_32.101998, latest
- 路径:
- **/details_harness|hendrycksTest-abstract_algebra|5_2024-03-05T00-51-32.101998.parquet
- **/details_harness|hendrycksTest-anatomy|5_2024-03-05T00-51-32.101998.parquet
- **/details_harness|hendrycksTest-astronomy|5_2024-03-05T00-51-32.101998.parquet
- **/details_harness|hendrycksTest-business_ethics|5_2024-03-05T00-51-32.101998.parquet
- **/details_harness|hendrycksTest-clinical_knowledge|5_2024-03-05T00-51-32.101998.parquet
- **/details_harness|hendrycksTest-college_biology|5_2024-03-05T00-51-32.101998.parquet
- **/details_harness|hendrycksTest-college_chemistry|5_2024-03-05T00-51-32.101998.parquet
- **/details_harness|hendrycksTest-college_computer_science|5_2024-03-05T00-51-32.101998.parquet
- **/details_harness|hendrycksTest-college_mathematics|5_2024-03-05T00-51-32.101998.parquet
- **/details_harness|hendrycksTest-college_medicine|5_2024-03-05T00-51-32.101998.parquet
- **/details_harness|hendrycksTest-college_physics|5_2024-03-05T00-51-32.101998.parquet
- **/details_harness|hendrycksTest-computer_security|5_2024-03-05T00-51-32.101998.parquet
- **/details_harness|hendrycksTest-conceptual_physics|5_2024-03-05T00-51-32.101998.parquet
- **/details_harness|hendrycksTest-econometrics|5_2024-03-05T00-51-32.101998.parquet
- **/details_harness|hendrycksTest-electrical_engineering|5_2024-03-05T00-51-32.101998.parquet
- **/details_harness|hendrycksTest-elementary_mathematics|5_2024-03-05T00-51-32.101998.parquet
- **/details_harness|hendrycksTest-formal_logic|5_2024-03-05T00-51-32.101998.parquet
- **/details_harness|hendrycksTest-global_facts|5_2024-03-05T00-51-32.101998.parquet
- **/details_harness|hendrycksTest-high_school_biology|5_2024-03-05T00-51-32.101998.parquet
- **/details_harness|hendrycksTest-high_school_chemistry|5_2024-03-05T00-51-32.101998.parquet
- **/details_harness|hendrycksTest-high_school_computer_science|5_2024-03-05T00-51-32.101998.parquet
- **/details_harness|hendrycksTest-high_school_european_history|5_2024-03-05T00-51-32.101998.parquet
- **/details_harness|hendrycksTest-high_school_geography|5_2024-03-05T00-51-32.101998.parquet
- **/details_harness|hendrycksTest-high_school_government_and_politics|5_2024-03-05T00-51-32.101998.parquet
- **/details_harness|hendrycksTest-high_school_macroeconomics|5_2024-03-05T00-51-32.101998.parquet
- **/details_harness|hendrycksTest-high_school_mathematics|5_2024-03-05T00-51-32.101998.parquet
- **/details_harness|hendrycksTest-high_school_microeconomics|5_2024-03-05T00-51-32.101998.parquet
- **/details_harness|hendrycksTest-high_school_physics|5_2024-03-05T00-51-32.101998.parquet
- **/details_harness|hendrycksTest-high_school_psychology|5_2024-03-05T00-51-32.101998.parquet
- **/details_harness|hendrycksTest-high_school_statistics|5_2024-03-05T00-51-32.101998.parquet
- **/details_harness|hendrycksTest-high_school_us_history|5_2024-03-05T00-51-32.101998.parquet
- **/details_harness|hendrycksTest-high_school_world_history|5_2024-03-05T00-51-32.101998.parquet
- **/details_harness|hendrycksTest-human_aging|5_2024-03-05T00-51-32.101998.parquet
- **/details_harness|hendrycksTest-human_sexuality|5_2024-03-05T00-51-32.101998.parquet
- **/details_harness|hendrycksTest-international_law|5_2024-03-05T00-51-32.101998.parquet
- **/details_harness|hendrycksTest-jurisprudence|5_2024-03-05T00-51-32.101998.parquet
- **/details_harness|hendrycksTest-logical_fallacies|5_2024-03-05T00-51-32.101998.parquet
- **/details_harness|hendrycksTest-machine_learning|5_2024-03-05T00-51-32.101998.parquet
- **/details_harness|hendrycksTest-management|5_2024-03-05T00-51-32.101998.parquet
- **/details_harness|hendrycksTest-marketing|5_2024-03-05T00-51-32.101998.parquet
- **/details_harness|hendrycksTest-medical_genetics|5_2024-03-05T00-51-32.101998.parquet
- **/details_harness|hendrycksTest-miscellaneous|5_2024-03-05T00-51-32.101998.parquet
- **/details_harness|hendrycksTest-moral_disputes|5_2024-03-05T00-51-32.101998.parquet
- **/details_harness|hendrycksTest-moral_scenarios|5_2024-03-05T00-51-32.101998.parquet
- **/details_harness|hendrycksTest-nutrition|5_2024-03-05T00-51-32.101998.parquet
- **/details_harness|hendrycksTest-philosophy|5_2024-03-05T00-51-32.101998.parquet
- **/details_harness|hendrycksTest-prehistory|5_2024-03-05T00-51-32.101998.parquet
- **/details_harness|hendrycksTest-professional_accounting|5_2024-03-05T00-51-32.101998.parquet
- **/details_harness|hendrycksTest-professional_law|5_2024-03-05T00-51-32.101998.parquet
- **/details_harness|hendrycksTest-professional_medicine|5_2024-03-05T00-51-32.101998.parquet
- **/details_harness|hendrycksTest-professional_psychology|5_2024-03-05T00-51-32.101998.parquet
- **/details_harness|hendrycksTest-public_relations|5_2024-03-05T00-51-32.101998.parquet
- **/details_harness|hendrycksTest-security_studies|5_2024-03-05T00-51-32.101998



