open-llm-leaderboard-old/details_habanoz__TinyLlama-1.1B-2T-lr-2e-4-3ep-dolly-15k-instruct-v1

该数据集是在评估模型habanoz/TinyLlama-1.1B-2T-lr-2e-4-3ep-dolly-15k-instruct-v1时自动生成的，包含1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，每次运行的结果以时间戳命名的分割形式存储，并且有一个额外的配置results用于存储所有运行的聚合结果。

Dataset automatically created during the evaluation run of model [habanoz/TinyLlama-1.1B-2T-lr-2e-4-3ep-dolly-15k-instruct-v1](https://huggingface.co/habanoz/TinyLlama-1.1B-2T-lr-2e-4-3ep-dolly-15k-instruct-v1) on the [Open LLM Leaderboard](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard). The dataset is composed of 1 configuration, each one corresponding to one of the evaluated task. The dataset has been created from 1 run(s). Each run can be found as a specific split in each configuration, the split being named using the timestamp of the run. The train split is always pointing to the latest results. An additional configuration results store all the aggregated results of the run (and is used to compute and display the aggregated metrics on the [Open LLM Leaderboard](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)).

原始信息汇总

数据集卡片 for Evaluation run of habanoz/TinyLlama-1.1B-2T-lr-2e-4-3ep-dolly-15k-instruct-v1

数据集描述

数据集概述

数据集是在模型 habanoz/TinyLlama-1.1B-2T-lr-2e-4-3ep-dolly-15k-instruct-v1 在 Open LLM Leaderboard 上的评估运行期间自动创建的。

数据集由1个配置组成，每个配置对应一个评估任务。

数据集是从1次运行中创建的。每次运行可以在每个配置中作为一个特定的分割找到，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" 存储了所有运行结果的聚合（用于计算和显示 Open LLM Leaderboard 上的聚合指标）。

最新结果

以下是来自2023-12-02T14:51:09.635098运行的最新结果：

python { "all": { "acc": 0.0, "acc_stderr": 0.0 }, "harness|gsm8k|5": { "acc": 0.0, "acc_stderr": 0.0 } }

数据集结构

数据配置

• config_name: harness_gsm8k_5

  • split: 2023_12_02T14_51_09.635098
    • path: **/details_harness|gsm8k|5_2023-12-02T14-51-09.635098.parquet
  • split: latest
    • path: **/details_harness|gsm8k|5_2023-12-02T14-51-09.635098.parquet

• config_name: results

  • split: 2023_12_02T14_51_09.635098
    • path: results_2023-12-02T14-51-09.635098.parquet
  • split: latest
    • path: results_2023-12-02T14-51-09.635098.parquet