thomwolf HF staff commited on
Commit
08143e9
1 Parent(s): 25a1ba9

Upload README.md with huggingface_hub

Browse files
Files changed (1) hide show
  1. README.md +8 -6
README.md CHANGED
@@ -13,9 +13,11 @@ dataset_summary: "Dataset automatically created during the evaluation run of mod
13
  \nTo load the details from a run, you can for instance do the following:\n```python\n\
14
  from datasets import load_dataset\ndata = load_dataset(\"open-llm-leaderboard/details_Kiddyz__testlm-1-1\"\
15
  ,\n\t\"harness_truthfulqa_mc_0\",\n\tsplit=\"train\")\n```\n\n## Latest results\n\
16
- \nThese are the [latest results from run 2023-08-16T10:49:10.911062](https://huggingface.co/datasets/open-llm-leaderboard/details_Kiddyz__testlm-1-1/blob/main/results_2023-08-16T10%3A49%3A10.911062.json):\n\
17
- \n```python\n{\n \"all\": {\n \"acc\": 0.5128834307003443,\n \"\
18
- acc_stderr\": 0.03501260490290392,\n \"acc_norm\": 0.5166256154161327,\n\
 
 
19
  \ \"acc_norm_stderr\": 0.03500071412093006,\n \"mc1\": 0.32802937576499386,\n\
20
  \ \"mc1_stderr\": 0.01643563293281503,\n \"mc2\": 0.48413168566081527,\n\
21
  \ \"mc2_stderr\": 0.015167638286466481\n },\n \"harness|arc:challenge|25\"\
@@ -197,7 +199,7 @@ configs:
197
  - split: latest
198
  path:
199
  - '**/details_harness|hellaswag|10_2023-08-16T10:49:10.911062.parquet'
200
- - config_name: harness_hendrycksTest
201
  data_files:
202
  - split: 2023_08_16T10_49_10.911062
203
  path:
@@ -783,7 +785,7 @@ configs:
783
  - '**/details_harness|truthfulqa:mc|0_2023-08-16T10:49:10.911062.parquet'
784
  - config_name: results
785
  data_files:
786
- - split: 2023_08_16T10_49_10.911062.parquet
787
  path:
788
  - results_2023-08-16T10:49:10.911062.parquet
789
  - split: latest
@@ -821,7 +823,7 @@ data = load_dataset("open-llm-leaderboard/details_Kiddyz__testlm-1-1",
821
 
822
  ## Latest results
823
 
824
- These are the [latest results from run 2023-08-16T10:49:10.911062](https://huggingface.co/datasets/open-llm-leaderboard/details_Kiddyz__testlm-1-1/blob/main/results_2023-08-16T10%3A49%3A10.911062.json):
825
 
826
  ```python
827
  {
 
13
  \nTo load the details from a run, you can for instance do the following:\n```python\n\
14
  from datasets import load_dataset\ndata = load_dataset(\"open-llm-leaderboard/details_Kiddyz__testlm-1-1\"\
15
  ,\n\t\"harness_truthfulqa_mc_0\",\n\tsplit=\"train\")\n```\n\n## Latest results\n\
16
+ \nThese are the [latest results from run 2023-08-16T10:49:10.911062](https://huggingface.co/datasets/open-llm-leaderboard/details_Kiddyz__testlm-1-1/blob/main/results_2023-08-16T10%3A49%3A10.911062.json)\
17
+ \ (note that their might be results for other tasks in the repos if successive evals\
18
+ \ didn't cover the same tasks. You find each in the results and the \"latest\" split\
19
+ \ for each eval):\n\n```python\n{\n \"all\": {\n \"acc\": 0.5128834307003443,\n\
20
+ \ \"acc_stderr\": 0.03501260490290392,\n \"acc_norm\": 0.5166256154161327,\n\
21
  \ \"acc_norm_stderr\": 0.03500071412093006,\n \"mc1\": 0.32802937576499386,\n\
22
  \ \"mc1_stderr\": 0.01643563293281503,\n \"mc2\": 0.48413168566081527,\n\
23
  \ \"mc2_stderr\": 0.015167638286466481\n },\n \"harness|arc:challenge|25\"\
 
199
  - split: latest
200
  path:
201
  - '**/details_harness|hellaswag|10_2023-08-16T10:49:10.911062.parquet'
202
+ - config_name: harness_hendrycksTest_5
203
  data_files:
204
  - split: 2023_08_16T10_49_10.911062
205
  path:
 
785
  - '**/details_harness|truthfulqa:mc|0_2023-08-16T10:49:10.911062.parquet'
786
  - config_name: results
787
  data_files:
788
+ - split: 2023_08_16T10_49_10.911062
789
  path:
790
  - results_2023-08-16T10:49:10.911062.parquet
791
  - split: latest
 
823
 
824
  ## Latest results
825
 
826
+ These are the [latest results from run 2023-08-16T10:49:10.911062](https://huggingface.co/datasets/open-llm-leaderboard/details_Kiddyz__testlm-1-1/blob/main/results_2023-08-16T10%3A49%3A10.911062.json) (note that their might be results for other tasks in the repos if successive evals didn't cover the same tasks. You find each in the results and the "latest" split for each eval):
827
 
828
  ```python
829
  {