SaylorTwift HF staff commited on
Commit
1ab856a
1 Parent(s): bb765e3

Upload README.md with huggingface_hub

Browse files
Files changed (1) hide show
  1. README.md +66 -539
README.md CHANGED
@@ -3,8 +3,8 @@ pretty_name: Evaluation run of Undi95/ReMM-v2.2-L2-13B
3
  dataset_summary: "Dataset automatically created during the evaluation run of model\
4
  \ [Undi95/ReMM-v2.2-L2-13B](https://huggingface.co/Undi95/ReMM-v2.2-L2-13B) on the\
5
  \ [Open LLM Leaderboard](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard).\n\
6
- \nThe dataset is composed of 61 configuration, each one coresponding to one of the\
7
- \ evaluated task.\n\nThe dataset has been created from 1 run(s). Each run can be\
8
  \ found as a specific split in each configuration, the split being named using the\
9
  \ timestamp of the run.The \"train\" split is always pointing to the latest results.\n\
10
  \nAn additional configuration \"results\" store all the aggregated results of the\
@@ -12,172 +12,20 @@ dataset_summary: "Dataset automatically created during the evaluation run of mod
12
  \ Leaderboard](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)).\n\
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_Undi95__ReMM-v2.2-L2-13B\"\
15
- ,\n\t\"harness_truthfulqa_mc_0\",\n\tsplit=\"train\")\n```\n\n## Latest results\n\
16
- \nThese are the [latest results from run 2023-10-03T16:45:21.105610](https://huggingface.co/datasets/open-llm-leaderboard/details_Undi95__ReMM-v2.2-L2-13B/blob/main/results_2023-10-03T16-45-21.105610.json)(note\
17
  \ that their might be results for other tasks in the repos if successive evals didn't\
18
  \ cover the same tasks. You find each in the results and the \"latest\" split for\
19
- \ each eval):\n\n```python\n{\n \"all\": {\n \"acc\": 0.564002539739021,\n\
20
- \ \"acc_stderr\": 0.034462415944422044,\n \"acc_norm\": 0.567760182755492,\n\
21
- \ \"acc_norm_stderr\": 0.034440565206339785,\n \"mc1\": 0.36964504283965727,\n\
22
- \ \"mc1_stderr\": 0.016898180706973888,\n \"mc2\": 0.5135116682163505,\n\
23
- \ \"mc2_stderr\": 0.015657648011440012\n },\n \"harness|arc:challenge|25\"\
24
- : {\n \"acc\": 0.5836177474402731,\n \"acc_stderr\": 0.01440561827943618,\n\
25
- \ \"acc_norm\": 0.6126279863481229,\n \"acc_norm_stderr\": 0.014235872487909869\n\
26
- \ },\n \"harness|hellaswag|10\": {\n \"acc\": 0.6488747261501693,\n\
27
- \ \"acc_stderr\": 0.004763465139038559,\n \"acc_norm\": 0.8415654252141008,\n\
28
- \ \"acc_norm_stderr\": 0.003644017383711596\n },\n \"harness|hendrycksTest-abstract_algebra|5\"\
29
- : {\n \"acc\": 0.37,\n \"acc_stderr\": 0.04852365870939099,\n \
30
- \ \"acc_norm\": 0.37,\n \"acc_norm_stderr\": 0.04852365870939099\n \
31
- \ },\n \"harness|hendrycksTest-anatomy|5\": {\n \"acc\": 0.5111111111111111,\n\
32
- \ \"acc_stderr\": 0.04318275491977976,\n \"acc_norm\": 0.5111111111111111,\n\
33
- \ \"acc_norm_stderr\": 0.04318275491977976\n },\n \"harness|hendrycksTest-astronomy|5\"\
34
- : {\n \"acc\": 0.5328947368421053,\n \"acc_stderr\": 0.040601270352363966,\n\
35
- \ \"acc_norm\": 0.5328947368421053,\n \"acc_norm_stderr\": 0.040601270352363966\n\
36
- \ },\n \"harness|hendrycksTest-business_ethics|5\": {\n \"acc\": 0.56,\n\
37
- \ \"acc_stderr\": 0.04988876515698589,\n \"acc_norm\": 0.56,\n \
38
- \ \"acc_norm_stderr\": 0.04988876515698589\n },\n \"harness|hendrycksTest-clinical_knowledge|5\"\
39
- : {\n \"acc\": 0.5924528301886792,\n \"acc_stderr\": 0.030242233800854494,\n\
40
- \ \"acc_norm\": 0.5924528301886792,\n \"acc_norm_stderr\": 0.030242233800854494\n\
41
- \ },\n \"harness|hendrycksTest-college_biology|5\": {\n \"acc\": 0.6041666666666666,\n\
42
- \ \"acc_stderr\": 0.04089465449325582,\n \"acc_norm\": 0.6041666666666666,\n\
43
- \ \"acc_norm_stderr\": 0.04089465449325582\n },\n \"harness|hendrycksTest-college_chemistry|5\"\
44
- : {\n \"acc\": 0.34,\n \"acc_stderr\": 0.04760952285695235,\n \
45
- \ \"acc_norm\": 0.34,\n \"acc_norm_stderr\": 0.04760952285695235\n \
46
- \ },\n \"harness|hendrycksTest-college_computer_science|5\": {\n \"acc\"\
47
- : 0.44,\n \"acc_stderr\": 0.04988876515698589,\n \"acc_norm\": 0.44,\n\
48
- \ \"acc_norm_stderr\": 0.04988876515698589\n },\n \"harness|hendrycksTest-college_mathematics|5\"\
49
- : {\n \"acc\": 0.35,\n \"acc_stderr\": 0.047937248544110196,\n \
50
- \ \"acc_norm\": 0.35,\n \"acc_norm_stderr\": 0.047937248544110196\n \
51
- \ },\n \"harness|hendrycksTest-college_medicine|5\": {\n \"acc\": 0.5317919075144508,\n\
52
- \ \"acc_stderr\": 0.03804749744364764,\n \"acc_norm\": 0.5317919075144508,\n\
53
- \ \"acc_norm_stderr\": 0.03804749744364764\n },\n \"harness|hendrycksTest-college_physics|5\"\
54
- : {\n \"acc\": 0.24509803921568626,\n \"acc_stderr\": 0.04280105837364397,\n\
55
- \ \"acc_norm\": 0.24509803921568626,\n \"acc_norm_stderr\": 0.04280105837364397\n\
56
- \ },\n \"harness|hendrycksTest-computer_security|5\": {\n \"acc\":\
57
- \ 0.71,\n \"acc_stderr\": 0.04560480215720685,\n \"acc_norm\": 0.71,\n\
58
- \ \"acc_norm_stderr\": 0.04560480215720685\n },\n \"harness|hendrycksTest-conceptual_physics|5\"\
59
- : {\n \"acc\": 0.4723404255319149,\n \"acc_stderr\": 0.03263597118409769,\n\
60
- \ \"acc_norm\": 0.4723404255319149,\n \"acc_norm_stderr\": 0.03263597118409769\n\
61
- \ },\n \"harness|hendrycksTest-econometrics|5\": {\n \"acc\": 0.2894736842105263,\n\
62
- \ \"acc_stderr\": 0.04266339443159394,\n \"acc_norm\": 0.2894736842105263,\n\
63
- \ \"acc_norm_stderr\": 0.04266339443159394\n },\n \"harness|hendrycksTest-electrical_engineering|5\"\
64
- : {\n \"acc\": 0.5172413793103449,\n \"acc_stderr\": 0.04164188720169375,\n\
65
- \ \"acc_norm\": 0.5172413793103449,\n \"acc_norm_stderr\": 0.04164188720169375\n\
66
- \ },\n \"harness|hendrycksTest-elementary_mathematics|5\": {\n \"acc\"\
67
- : 0.3439153439153439,\n \"acc_stderr\": 0.024464426625596433,\n \"\
68
- acc_norm\": 0.3439153439153439,\n \"acc_norm_stderr\": 0.024464426625596433\n\
69
- \ },\n \"harness|hendrycksTest-formal_logic|5\": {\n \"acc\": 0.3492063492063492,\n\
70
- \ \"acc_stderr\": 0.04263906892795132,\n \"acc_norm\": 0.3492063492063492,\n\
71
- \ \"acc_norm_stderr\": 0.04263906892795132\n },\n \"harness|hendrycksTest-global_facts|5\"\
72
- : {\n \"acc\": 0.38,\n \"acc_stderr\": 0.04878317312145633,\n \
73
- \ \"acc_norm\": 0.38,\n \"acc_norm_stderr\": 0.04878317312145633\n \
74
- \ },\n \"harness|hendrycksTest-high_school_biology|5\": {\n \"acc\": 0.6483870967741936,\n\
75
- \ \"acc_stderr\": 0.02716253782694846,\n \"acc_norm\": 0.6483870967741936,\n\
76
- \ \"acc_norm_stderr\": 0.02716253782694846\n },\n \"harness|hendrycksTest-high_school_chemistry|5\"\
77
- : {\n \"acc\": 0.43842364532019706,\n \"acc_stderr\": 0.03491207857486518,\n\
78
- \ \"acc_norm\": 0.43842364532019706,\n \"acc_norm_stderr\": 0.03491207857486518\n\
79
- \ },\n \"harness|hendrycksTest-high_school_computer_science|5\": {\n \
80
- \ \"acc\": 0.56,\n \"acc_stderr\": 0.04988876515698589,\n \"acc_norm\"\
81
- : 0.56,\n \"acc_norm_stderr\": 0.04988876515698589\n },\n \"harness|hendrycksTest-high_school_european_history|5\"\
82
- : {\n \"acc\": 0.6666666666666666,\n \"acc_stderr\": 0.0368105086916155,\n\
83
- \ \"acc_norm\": 0.6666666666666666,\n \"acc_norm_stderr\": 0.0368105086916155\n\
84
- \ },\n \"harness|hendrycksTest-high_school_geography|5\": {\n \"acc\"\
85
- : 0.702020202020202,\n \"acc_stderr\": 0.03258630383836556,\n \"acc_norm\"\
86
- : 0.702020202020202,\n \"acc_norm_stderr\": 0.03258630383836556\n },\n\
87
- \ \"harness|hendrycksTest-high_school_government_and_politics|5\": {\n \
88
- \ \"acc\": 0.7979274611398963,\n \"acc_stderr\": 0.028979089794296732,\n\
89
- \ \"acc_norm\": 0.7979274611398963,\n \"acc_norm_stderr\": 0.028979089794296732\n\
90
- \ },\n \"harness|hendrycksTest-high_school_macroeconomics|5\": {\n \
91
- \ \"acc\": 0.5256410256410257,\n \"acc_stderr\": 0.025317649726448663,\n\
92
- \ \"acc_norm\": 0.5256410256410257,\n \"acc_norm_stderr\": 0.025317649726448663\n\
93
- \ },\n \"harness|hendrycksTest-high_school_mathematics|5\": {\n \"\
94
- acc\": 0.32592592592592595,\n \"acc_stderr\": 0.028578348365473072,\n \
95
- \ \"acc_norm\": 0.32592592592592595,\n \"acc_norm_stderr\": 0.028578348365473072\n\
96
- \ },\n \"harness|hendrycksTest-high_school_microeconomics|5\": {\n \
97
- \ \"acc\": 0.5966386554621849,\n \"acc_stderr\": 0.031866081214088314,\n\
98
- \ \"acc_norm\": 0.5966386554621849,\n \"acc_norm_stderr\": 0.031866081214088314\n\
99
- \ },\n \"harness|hendrycksTest-high_school_physics|5\": {\n \"acc\"\
100
- : 0.3509933774834437,\n \"acc_stderr\": 0.03896981964257375,\n \"\
101
- acc_norm\": 0.3509933774834437,\n \"acc_norm_stderr\": 0.03896981964257375\n\
102
- \ },\n \"harness|hendrycksTest-high_school_psychology|5\": {\n \"acc\"\
103
- : 0.7302752293577982,\n \"acc_stderr\": 0.01902848671111544,\n \"\
104
- acc_norm\": 0.7302752293577982,\n \"acc_norm_stderr\": 0.01902848671111544\n\
105
- \ },\n \"harness|hendrycksTest-high_school_statistics|5\": {\n \"acc\"\
106
- : 0.41203703703703703,\n \"acc_stderr\": 0.03356787758160835,\n \"\
107
- acc_norm\": 0.41203703703703703,\n \"acc_norm_stderr\": 0.03356787758160835\n\
108
- \ },\n \"harness|hendrycksTest-high_school_us_history|5\": {\n \"acc\"\
109
- : 0.7647058823529411,\n \"acc_stderr\": 0.029771775228145628,\n \"\
110
- acc_norm\": 0.7647058823529411,\n \"acc_norm_stderr\": 0.029771775228145628\n\
111
- \ },\n \"harness|hendrycksTest-high_school_world_history|5\": {\n \"\
112
- acc\": 0.7637130801687764,\n \"acc_stderr\": 0.02765215314415927,\n \
113
- \ \"acc_norm\": 0.7637130801687764,\n \"acc_norm_stderr\": 0.02765215314415927\n\
114
- \ },\n \"harness|hendrycksTest-human_aging|5\": {\n \"acc\": 0.6771300448430493,\n\
115
- \ \"acc_stderr\": 0.03138147637575499,\n \"acc_norm\": 0.6771300448430493,\n\
116
- \ \"acc_norm_stderr\": 0.03138147637575499\n },\n \"harness|hendrycksTest-human_sexuality|5\"\
117
- : {\n \"acc\": 0.6183206106870229,\n \"acc_stderr\": 0.04260735157644559,\n\
118
- \ \"acc_norm\": 0.6183206106870229,\n \"acc_norm_stderr\": 0.04260735157644559\n\
119
- \ },\n \"harness|hendrycksTest-international_law|5\": {\n \"acc\":\
120
- \ 0.7272727272727273,\n \"acc_stderr\": 0.04065578140908705,\n \"\
121
- acc_norm\": 0.7272727272727273,\n \"acc_norm_stderr\": 0.04065578140908705\n\
122
- \ },\n \"harness|hendrycksTest-jurisprudence|5\": {\n \"acc\": 0.75,\n\
123
- \ \"acc_stderr\": 0.04186091791394607,\n \"acc_norm\": 0.75,\n \
124
- \ \"acc_norm_stderr\": 0.04186091791394607\n },\n \"harness|hendrycksTest-logical_fallacies|5\"\
125
- : {\n \"acc\": 0.6993865030674846,\n \"acc_stderr\": 0.03602511318806771,\n\
126
- \ \"acc_norm\": 0.6993865030674846,\n \"acc_norm_stderr\": 0.03602511318806771\n\
127
- \ },\n \"harness|hendrycksTest-machine_learning|5\": {\n \"acc\": 0.36607142857142855,\n\
128
- \ \"acc_stderr\": 0.0457237235873743,\n \"acc_norm\": 0.36607142857142855,\n\
129
- \ \"acc_norm_stderr\": 0.0457237235873743\n },\n \"harness|hendrycksTest-management|5\"\
130
- : {\n \"acc\": 0.7087378640776699,\n \"acc_stderr\": 0.044986763205729224,\n\
131
- \ \"acc_norm\": 0.7087378640776699,\n \"acc_norm_stderr\": 0.044986763205729224\n\
132
- \ },\n \"harness|hendrycksTest-marketing|5\": {\n \"acc\": 0.7948717948717948,\n\
133
- \ \"acc_stderr\": 0.026453508054040332,\n \"acc_norm\": 0.7948717948717948,\n\
134
- \ \"acc_norm_stderr\": 0.026453508054040332\n },\n \"harness|hendrycksTest-medical_genetics|5\"\
135
- : {\n \"acc\": 0.59,\n \"acc_stderr\": 0.049431107042371025,\n \
136
- \ \"acc_norm\": 0.59,\n \"acc_norm_stderr\": 0.049431107042371025\n \
137
- \ },\n \"harness|hendrycksTest-miscellaneous|5\": {\n \"acc\": 0.7611749680715197,\n\
138
- \ \"acc_stderr\": 0.015246803197398675,\n \"acc_norm\": 0.7611749680715197,\n\
139
- \ \"acc_norm_stderr\": 0.015246803197398675\n },\n \"harness|hendrycksTest-moral_disputes|5\"\
140
- : {\n \"acc\": 0.6358381502890174,\n \"acc_stderr\": 0.025906632631016127,\n\
141
- \ \"acc_norm\": 0.6358381502890174,\n \"acc_norm_stderr\": 0.025906632631016127\n\
142
- \ },\n \"harness|hendrycksTest-moral_scenarios|5\": {\n \"acc\": 0.45363128491620114,\n\
143
- \ \"acc_stderr\": 0.016650437588269073,\n \"acc_norm\": 0.45363128491620114,\n\
144
- \ \"acc_norm_stderr\": 0.016650437588269073\n },\n \"harness|hendrycksTest-nutrition|5\"\
145
- : {\n \"acc\": 0.6143790849673203,\n \"acc_stderr\": 0.02787074527829028,\n\
146
- \ \"acc_norm\": 0.6143790849673203,\n \"acc_norm_stderr\": 0.02787074527829028\n\
147
- \ },\n \"harness|hendrycksTest-philosophy|5\": {\n \"acc\": 0.6430868167202572,\n\
148
- \ \"acc_stderr\": 0.027210420375934023,\n \"acc_norm\": 0.6430868167202572,\n\
149
- \ \"acc_norm_stderr\": 0.027210420375934023\n },\n \"harness|hendrycksTest-prehistory|5\"\
150
- : {\n \"acc\": 0.6203703703703703,\n \"acc_stderr\": 0.027002521034516478,\n\
151
- \ \"acc_norm\": 0.6203703703703703,\n \"acc_norm_stderr\": 0.027002521034516478\n\
152
- \ },\n \"harness|hendrycksTest-professional_accounting|5\": {\n \"\
153
- acc\": 0.425531914893617,\n \"acc_stderr\": 0.029494827600144373,\n \
154
- \ \"acc_norm\": 0.425531914893617,\n \"acc_norm_stderr\": 0.029494827600144373\n\
155
- \ },\n \"harness|hendrycksTest-professional_law|5\": {\n \"acc\": 0.42503259452411996,\n\
156
- \ \"acc_stderr\": 0.012625879884891998,\n \"acc_norm\": 0.42503259452411996,\n\
157
- \ \"acc_norm_stderr\": 0.012625879884891998\n },\n \"harness|hendrycksTest-professional_medicine|5\"\
158
- : {\n \"acc\": 0.5183823529411765,\n \"acc_stderr\": 0.030352303395351964,\n\
159
- \ \"acc_norm\": 0.5183823529411765,\n \"acc_norm_stderr\": 0.030352303395351964\n\
160
- \ },\n \"harness|hendrycksTest-professional_psychology|5\": {\n \"\
161
- acc\": 0.5816993464052288,\n \"acc_stderr\": 0.019955975145835546,\n \
162
- \ \"acc_norm\": 0.5816993464052288,\n \"acc_norm_stderr\": 0.019955975145835546\n\
163
- \ },\n \"harness|hendrycksTest-public_relations|5\": {\n \"acc\": 0.6545454545454545,\n\
164
- \ \"acc_stderr\": 0.04554619617541054,\n \"acc_norm\": 0.6545454545454545,\n\
165
- \ \"acc_norm_stderr\": 0.04554619617541054\n },\n \"harness|hendrycksTest-security_studies|5\"\
166
- : {\n \"acc\": 0.6244897959183674,\n \"acc_stderr\": 0.03100120903989484,\n\
167
- \ \"acc_norm\": 0.6244897959183674,\n \"acc_norm_stderr\": 0.03100120903989484\n\
168
- \ },\n \"harness|hendrycksTest-sociology|5\": {\n \"acc\": 0.6865671641791045,\n\
169
- \ \"acc_stderr\": 0.032801882053486435,\n \"acc_norm\": 0.6865671641791045,\n\
170
- \ \"acc_norm_stderr\": 0.032801882053486435\n },\n \"harness|hendrycksTest-us_foreign_policy|5\"\
171
- : {\n \"acc\": 0.82,\n \"acc_stderr\": 0.038612291966536934,\n \
172
- \ \"acc_norm\": 0.82,\n \"acc_norm_stderr\": 0.038612291966536934\n \
173
- \ },\n \"harness|hendrycksTest-virology|5\": {\n \"acc\": 0.463855421686747,\n\
174
- \ \"acc_stderr\": 0.03882310850890594,\n \"acc_norm\": 0.463855421686747,\n\
175
- \ \"acc_norm_stderr\": 0.03882310850890594\n },\n \"harness|hendrycksTest-world_religions|5\"\
176
- : {\n \"acc\": 0.7602339181286549,\n \"acc_stderr\": 0.03274485211946956,\n\
177
- \ \"acc_norm\": 0.7602339181286549,\n \"acc_norm_stderr\": 0.03274485211946956\n\
178
- \ },\n \"harness|truthfulqa:mc|0\": {\n \"mc1\": 0.36964504283965727,\n\
179
- \ \"mc1_stderr\": 0.016898180706973888,\n \"mc2\": 0.5135116682163505,\n\
180
- \ \"mc2_stderr\": 0.015657648011440012\n }\n}\n```"
181
  repo_url: https://huggingface.co/Undi95/ReMM-v2.2-L2-13B
182
  leaderboard_url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard
183
  point_of_contact: clementine@hf.co
@@ -190,6 +38,22 @@ configs:
190
  - split: latest
191
  path:
192
  - '**/details_harness|arc:challenge|25_2023-10-03T16-45-21.105610.parquet'
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
193
  - config_name: harness_hellaswag_10
194
  data_files:
195
  - split: 2023_10_03T16_45_21.105610
@@ -782,14 +646,25 @@ configs:
782
  - split: latest
783
  path:
784
  - '**/details_harness|truthfulqa:mc|0_2023-10-03T16-45-21.105610.parquet'
 
 
 
 
 
 
 
 
785
  - config_name: results
786
  data_files:
787
  - split: 2023_10_03T16_45_21.105610
788
  path:
789
  - results_2023-10-03T16-45-21.105610.parquet
 
 
 
790
  - split: latest
791
  path:
792
- - results_2023-10-03T16-45-21.105610.parquet
793
  ---
794
 
795
  # Dataset Card for Evaluation run of Undi95/ReMM-v2.2-L2-13B
@@ -806,9 +681,9 @@ configs:
806
 
807
  Dataset automatically created during the evaluation run of model [Undi95/ReMM-v2.2-L2-13B](https://huggingface.co/Undi95/ReMM-v2.2-L2-13B) on the [Open LLM Leaderboard](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard).
808
 
809
- The dataset is composed of 61 configuration, each one coresponding to one of the evaluated task.
810
 
811
- 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.
812
 
813
  An additional configuration "results" store all the aggregated results of the run (and is used to compute and display the agregated metrics on the [Open LLM Leaderboard](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)).
814
 
@@ -816,385 +691,37 @@ To load the details from a run, you can for instance do the following:
816
  ```python
817
  from datasets import load_dataset
818
  data = load_dataset("open-llm-leaderboard/details_Undi95__ReMM-v2.2-L2-13B",
819
- "harness_truthfulqa_mc_0",
820
  split="train")
821
  ```
822
 
823
  ## Latest results
824
 
825
- These are the [latest results from run 2023-10-03T16:45:21.105610](https://huggingface.co/datasets/open-llm-leaderboard/details_Undi95__ReMM-v2.2-L2-13B/blob/main/results_2023-10-03T16-45-21.105610.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):
826
 
827
  ```python
828
  {
829
  "all": {
830
- "acc": 0.564002539739021,
831
- "acc_stderr": 0.034462415944422044,
832
- "acc_norm": 0.567760182755492,
833
- "acc_norm_stderr": 0.034440565206339785,
834
- "mc1": 0.36964504283965727,
835
- "mc1_stderr": 0.016898180706973888,
836
- "mc2": 0.5135116682163505,
837
- "mc2_stderr": 0.015657648011440012
838
- },
839
- "harness|arc:challenge|25": {
840
- "acc": 0.5836177474402731,
841
- "acc_stderr": 0.01440561827943618,
842
- "acc_norm": 0.6126279863481229,
843
- "acc_norm_stderr": 0.014235872487909869
844
- },
845
- "harness|hellaswag|10": {
846
- "acc": 0.6488747261501693,
847
- "acc_stderr": 0.004763465139038559,
848
- "acc_norm": 0.8415654252141008,
849
- "acc_norm_stderr": 0.003644017383711596
850
- },
851
- "harness|hendrycksTest-abstract_algebra|5": {
852
- "acc": 0.37,
853
- "acc_stderr": 0.04852365870939099,
854
- "acc_norm": 0.37,
855
- "acc_norm_stderr": 0.04852365870939099
856
- },
857
- "harness|hendrycksTest-anatomy|5": {
858
- "acc": 0.5111111111111111,
859
- "acc_stderr": 0.04318275491977976,
860
- "acc_norm": 0.5111111111111111,
861
- "acc_norm_stderr": 0.04318275491977976
862
- },
863
- "harness|hendrycksTest-astronomy|5": {
864
- "acc": 0.5328947368421053,
865
- "acc_stderr": 0.040601270352363966,
866
- "acc_norm": 0.5328947368421053,
867
- "acc_norm_stderr": 0.040601270352363966
868
- },
869
- "harness|hendrycksTest-business_ethics|5": {
870
- "acc": 0.56,
871
- "acc_stderr": 0.04988876515698589,
872
- "acc_norm": 0.56,
873
- "acc_norm_stderr": 0.04988876515698589
874
- },
875
- "harness|hendrycksTest-clinical_knowledge|5": {
876
- "acc": 0.5924528301886792,
877
- "acc_stderr": 0.030242233800854494,
878
- "acc_norm": 0.5924528301886792,
879
- "acc_norm_stderr": 0.030242233800854494
880
- },
881
- "harness|hendrycksTest-college_biology|5": {
882
- "acc": 0.6041666666666666,
883
- "acc_stderr": 0.04089465449325582,
884
- "acc_norm": 0.6041666666666666,
885
- "acc_norm_stderr": 0.04089465449325582
886
- },
887
- "harness|hendrycksTest-college_chemistry|5": {
888
- "acc": 0.34,
889
- "acc_stderr": 0.04760952285695235,
890
- "acc_norm": 0.34,
891
- "acc_norm_stderr": 0.04760952285695235
892
- },
893
- "harness|hendrycksTest-college_computer_science|5": {
894
- "acc": 0.44,
895
- "acc_stderr": 0.04988876515698589,
896
- "acc_norm": 0.44,
897
- "acc_norm_stderr": 0.04988876515698589
898
- },
899
- "harness|hendrycksTest-college_mathematics|5": {
900
- "acc": 0.35,
901
- "acc_stderr": 0.047937248544110196,
902
- "acc_norm": 0.35,
903
- "acc_norm_stderr": 0.047937248544110196
904
- },
905
- "harness|hendrycksTest-college_medicine|5": {
906
- "acc": 0.5317919075144508,
907
- "acc_stderr": 0.03804749744364764,
908
- "acc_norm": 0.5317919075144508,
909
- "acc_norm_stderr": 0.03804749744364764
910
- },
911
- "harness|hendrycksTest-college_physics|5": {
912
- "acc": 0.24509803921568626,
913
- "acc_stderr": 0.04280105837364397,
914
- "acc_norm": 0.24509803921568626,
915
- "acc_norm_stderr": 0.04280105837364397
916
- },
917
- "harness|hendrycksTest-computer_security|5": {
918
- "acc": 0.71,
919
- "acc_stderr": 0.04560480215720685,
920
- "acc_norm": 0.71,
921
- "acc_norm_stderr": 0.04560480215720685
922
- },
923
- "harness|hendrycksTest-conceptual_physics|5": {
924
- "acc": 0.4723404255319149,
925
- "acc_stderr": 0.03263597118409769,
926
- "acc_norm": 0.4723404255319149,
927
- "acc_norm_stderr": 0.03263597118409769
928
- },
929
- "harness|hendrycksTest-econometrics|5": {
930
- "acc": 0.2894736842105263,
931
- "acc_stderr": 0.04266339443159394,
932
- "acc_norm": 0.2894736842105263,
933
- "acc_norm_stderr": 0.04266339443159394
934
- },
935
- "harness|hendrycksTest-electrical_engineering|5": {
936
- "acc": 0.5172413793103449,
937
- "acc_stderr": 0.04164188720169375,
938
- "acc_norm": 0.5172413793103449,
939
- "acc_norm_stderr": 0.04164188720169375
940
- },
941
- "harness|hendrycksTest-elementary_mathematics|5": {
942
- "acc": 0.3439153439153439,
943
- "acc_stderr": 0.024464426625596433,
944
- "acc_norm": 0.3439153439153439,
945
- "acc_norm_stderr": 0.024464426625596433
946
- },
947
- "harness|hendrycksTest-formal_logic|5": {
948
- "acc": 0.3492063492063492,
949
- "acc_stderr": 0.04263906892795132,
950
- "acc_norm": 0.3492063492063492,
951
- "acc_norm_stderr": 0.04263906892795132
952
- },
953
- "harness|hendrycksTest-global_facts|5": {
954
- "acc": 0.38,
955
- "acc_stderr": 0.04878317312145633,
956
- "acc_norm": 0.38,
957
- "acc_norm_stderr": 0.04878317312145633
958
- },
959
- "harness|hendrycksTest-high_school_biology|5": {
960
- "acc": 0.6483870967741936,
961
- "acc_stderr": 0.02716253782694846,
962
- "acc_norm": 0.6483870967741936,
963
- "acc_norm_stderr": 0.02716253782694846
964
- },
965
- "harness|hendrycksTest-high_school_chemistry|5": {
966
- "acc": 0.43842364532019706,
967
- "acc_stderr": 0.03491207857486518,
968
- "acc_norm": 0.43842364532019706,
969
- "acc_norm_stderr": 0.03491207857486518
970
- },
971
- "harness|hendrycksTest-high_school_computer_science|5": {
972
- "acc": 0.56,
973
- "acc_stderr": 0.04988876515698589,
974
- "acc_norm": 0.56,
975
- "acc_norm_stderr": 0.04988876515698589
976
- },
977
- "harness|hendrycksTest-high_school_european_history|5": {
978
- "acc": 0.6666666666666666,
979
- "acc_stderr": 0.0368105086916155,
980
- "acc_norm": 0.6666666666666666,
981
- "acc_norm_stderr": 0.0368105086916155
982
- },
983
- "harness|hendrycksTest-high_school_geography|5": {
984
- "acc": 0.702020202020202,
985
- "acc_stderr": 0.03258630383836556,
986
- "acc_norm": 0.702020202020202,
987
- "acc_norm_stderr": 0.03258630383836556
988
- },
989
- "harness|hendrycksTest-high_school_government_and_politics|5": {
990
- "acc": 0.7979274611398963,
991
- "acc_stderr": 0.028979089794296732,
992
- "acc_norm": 0.7979274611398963,
993
- "acc_norm_stderr": 0.028979089794296732
994
- },
995
- "harness|hendrycksTest-high_school_macroeconomics|5": {
996
- "acc": 0.5256410256410257,
997
- "acc_stderr": 0.025317649726448663,
998
- "acc_norm": 0.5256410256410257,
999
- "acc_norm_stderr": 0.025317649726448663
1000
- },
1001
- "harness|hendrycksTest-high_school_mathematics|5": {
1002
- "acc": 0.32592592592592595,
1003
- "acc_stderr": 0.028578348365473072,
1004
- "acc_norm": 0.32592592592592595,
1005
- "acc_norm_stderr": 0.028578348365473072
1006
- },
1007
- "harness|hendrycksTest-high_school_microeconomics|5": {
1008
- "acc": 0.5966386554621849,
1009
- "acc_stderr": 0.031866081214088314,
1010
- "acc_norm": 0.5966386554621849,
1011
- "acc_norm_stderr": 0.031866081214088314
1012
- },
1013
- "harness|hendrycksTest-high_school_physics|5": {
1014
- "acc": 0.3509933774834437,
1015
- "acc_stderr": 0.03896981964257375,
1016
- "acc_norm": 0.3509933774834437,
1017
- "acc_norm_stderr": 0.03896981964257375
1018
- },
1019
- "harness|hendrycksTest-high_school_psychology|5": {
1020
- "acc": 0.7302752293577982,
1021
- "acc_stderr": 0.01902848671111544,
1022
- "acc_norm": 0.7302752293577982,
1023
- "acc_norm_stderr": 0.01902848671111544
1024
- },
1025
- "harness|hendrycksTest-high_school_statistics|5": {
1026
- "acc": 0.41203703703703703,
1027
- "acc_stderr": 0.03356787758160835,
1028
- "acc_norm": 0.41203703703703703,
1029
- "acc_norm_stderr": 0.03356787758160835
1030
- },
1031
- "harness|hendrycksTest-high_school_us_history|5": {
1032
- "acc": 0.7647058823529411,
1033
- "acc_stderr": 0.029771775228145628,
1034
- "acc_norm": 0.7647058823529411,
1035
- "acc_norm_stderr": 0.029771775228145628
1036
- },
1037
- "harness|hendrycksTest-high_school_world_history|5": {
1038
- "acc": 0.7637130801687764,
1039
- "acc_stderr": 0.02765215314415927,
1040
- "acc_norm": 0.7637130801687764,
1041
- "acc_norm_stderr": 0.02765215314415927
1042
- },
1043
- "harness|hendrycksTest-human_aging|5": {
1044
- "acc": 0.6771300448430493,
1045
- "acc_stderr": 0.03138147637575499,
1046
- "acc_norm": 0.6771300448430493,
1047
- "acc_norm_stderr": 0.03138147637575499
1048
- },
1049
- "harness|hendrycksTest-human_sexuality|5": {
1050
- "acc": 0.6183206106870229,
1051
- "acc_stderr": 0.04260735157644559,
1052
- "acc_norm": 0.6183206106870229,
1053
- "acc_norm_stderr": 0.04260735157644559
1054
- },
1055
- "harness|hendrycksTest-international_law|5": {
1056
- "acc": 0.7272727272727273,
1057
- "acc_stderr": 0.04065578140908705,
1058
- "acc_norm": 0.7272727272727273,
1059
- "acc_norm_stderr": 0.04065578140908705
1060
- },
1061
- "harness|hendrycksTest-jurisprudence|5": {
1062
- "acc": 0.75,
1063
- "acc_stderr": 0.04186091791394607,
1064
- "acc_norm": 0.75,
1065
- "acc_norm_stderr": 0.04186091791394607
1066
- },
1067
- "harness|hendrycksTest-logical_fallacies|5": {
1068
- "acc": 0.6993865030674846,
1069
- "acc_stderr": 0.03602511318806771,
1070
- "acc_norm": 0.6993865030674846,
1071
- "acc_norm_stderr": 0.03602511318806771
1072
- },
1073
- "harness|hendrycksTest-machine_learning|5": {
1074
- "acc": 0.36607142857142855,
1075
- "acc_stderr": 0.0457237235873743,
1076
- "acc_norm": 0.36607142857142855,
1077
- "acc_norm_stderr": 0.0457237235873743
1078
- },
1079
- "harness|hendrycksTest-management|5": {
1080
- "acc": 0.7087378640776699,
1081
- "acc_stderr": 0.044986763205729224,
1082
- "acc_norm": 0.7087378640776699,
1083
- "acc_norm_stderr": 0.044986763205729224
1084
- },
1085
- "harness|hendrycksTest-marketing|5": {
1086
- "acc": 0.7948717948717948,
1087
- "acc_stderr": 0.026453508054040332,
1088
- "acc_norm": 0.7948717948717948,
1089
- "acc_norm_stderr": 0.026453508054040332
1090
- },
1091
- "harness|hendrycksTest-medical_genetics|5": {
1092
- "acc": 0.59,
1093
- "acc_stderr": 0.049431107042371025,
1094
- "acc_norm": 0.59,
1095
- "acc_norm_stderr": 0.049431107042371025
1096
- },
1097
- "harness|hendrycksTest-miscellaneous|5": {
1098
- "acc": 0.7611749680715197,
1099
- "acc_stderr": 0.015246803197398675,
1100
- "acc_norm": 0.7611749680715197,
1101
- "acc_norm_stderr": 0.015246803197398675
1102
- },
1103
- "harness|hendrycksTest-moral_disputes|5": {
1104
- "acc": 0.6358381502890174,
1105
- "acc_stderr": 0.025906632631016127,
1106
- "acc_norm": 0.6358381502890174,
1107
- "acc_norm_stderr": 0.025906632631016127
1108
- },
1109
- "harness|hendrycksTest-moral_scenarios|5": {
1110
- "acc": 0.45363128491620114,
1111
- "acc_stderr": 0.016650437588269073,
1112
- "acc_norm": 0.45363128491620114,
1113
- "acc_norm_stderr": 0.016650437588269073
1114
- },
1115
- "harness|hendrycksTest-nutrition|5": {
1116
- "acc": 0.6143790849673203,
1117
- "acc_stderr": 0.02787074527829028,
1118
- "acc_norm": 0.6143790849673203,
1119
- "acc_norm_stderr": 0.02787074527829028
1120
- },
1121
- "harness|hendrycksTest-philosophy|5": {
1122
- "acc": 0.6430868167202572,
1123
- "acc_stderr": 0.027210420375934023,
1124
- "acc_norm": 0.6430868167202572,
1125
- "acc_norm_stderr": 0.027210420375934023
1126
- },
1127
- "harness|hendrycksTest-prehistory|5": {
1128
- "acc": 0.6203703703703703,
1129
- "acc_stderr": 0.027002521034516478,
1130
- "acc_norm": 0.6203703703703703,
1131
- "acc_norm_stderr": 0.027002521034516478
1132
- },
1133
- "harness|hendrycksTest-professional_accounting|5": {
1134
- "acc": 0.425531914893617,
1135
- "acc_stderr": 0.029494827600144373,
1136
- "acc_norm": 0.425531914893617,
1137
- "acc_norm_stderr": 0.029494827600144373
1138
- },
1139
- "harness|hendrycksTest-professional_law|5": {
1140
- "acc": 0.42503259452411996,
1141
- "acc_stderr": 0.012625879884891998,
1142
- "acc_norm": 0.42503259452411996,
1143
- "acc_norm_stderr": 0.012625879884891998
1144
- },
1145
- "harness|hendrycksTest-professional_medicine|5": {
1146
- "acc": 0.5183823529411765,
1147
- "acc_stderr": 0.030352303395351964,
1148
- "acc_norm": 0.5183823529411765,
1149
- "acc_norm_stderr": 0.030352303395351964
1150
- },
1151
- "harness|hendrycksTest-professional_psychology|5": {
1152
- "acc": 0.5816993464052288,
1153
- "acc_stderr": 0.019955975145835546,
1154
- "acc_norm": 0.5816993464052288,
1155
- "acc_norm_stderr": 0.019955975145835546
1156
- },
1157
- "harness|hendrycksTest-public_relations|5": {
1158
- "acc": 0.6545454545454545,
1159
- "acc_stderr": 0.04554619617541054,
1160
- "acc_norm": 0.6545454545454545,
1161
- "acc_norm_stderr": 0.04554619617541054
1162
- },
1163
- "harness|hendrycksTest-security_studies|5": {
1164
- "acc": 0.6244897959183674,
1165
- "acc_stderr": 0.03100120903989484,
1166
- "acc_norm": 0.6244897959183674,
1167
- "acc_norm_stderr": 0.03100120903989484
1168
- },
1169
- "harness|hendrycksTest-sociology|5": {
1170
- "acc": 0.6865671641791045,
1171
- "acc_stderr": 0.032801882053486435,
1172
- "acc_norm": 0.6865671641791045,
1173
- "acc_norm_stderr": 0.032801882053486435
1174
- },
1175
- "harness|hendrycksTest-us_foreign_policy|5": {
1176
- "acc": 0.82,
1177
- "acc_stderr": 0.038612291966536934,
1178
- "acc_norm": 0.82,
1179
- "acc_norm_stderr": 0.038612291966536934
1180
- },
1181
- "harness|hendrycksTest-virology|5": {
1182
- "acc": 0.463855421686747,
1183
- "acc_stderr": 0.03882310850890594,
1184
- "acc_norm": 0.463855421686747,
1185
- "acc_norm_stderr": 0.03882310850890594
1186
- },
1187
- "harness|hendrycksTest-world_religions|5": {
1188
- "acc": 0.7602339181286549,
1189
- "acc_stderr": 0.03274485211946956,
1190
- "acc_norm": 0.7602339181286549,
1191
- "acc_norm_stderr": 0.03274485211946956
1192
- },
1193
- "harness|truthfulqa:mc|0": {
1194
- "mc1": 0.36964504283965727,
1195
- "mc1_stderr": 0.016898180706973888,
1196
- "mc2": 0.5135116682163505,
1197
- "mc2_stderr": 0.015657648011440012
1198
  }
1199
  }
1200
  ```
 
3
  dataset_summary: "Dataset automatically created during the evaluation run of model\
4
  \ [Undi95/ReMM-v2.2-L2-13B](https://huggingface.co/Undi95/ReMM-v2.2-L2-13B) on the\
5
  \ [Open LLM Leaderboard](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard).\n\
6
+ \nThe dataset is composed of 64 configuration, each one coresponding to one of the\
7
+ \ evaluated task.\n\nThe dataset has been created from 2 run(s). Each run can be\
8
  \ found as a specific split in each configuration, the split being named using the\
9
  \ timestamp of the run.The \"train\" split is always pointing to the latest results.\n\
10
  \nAn additional configuration \"results\" store all the aggregated results of the\
 
12
  \ Leaderboard](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)).\n\
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_Undi95__ReMM-v2.2-L2-13B\"\
15
+ ,\n\t\"harness_winogrande_5\",\n\tsplit=\"train\")\n```\n\n## Latest results\n\n\
16
+ These are the [latest results from run 2023-10-23T13:54:57.235808](https://huggingface.co/datasets/open-llm-leaderboard/details_Undi95__ReMM-v2.2-L2-13B/blob/main/results_2023-10-23T13-54-57.235808.json)(note\
17
  \ that their might be results for other tasks in the repos if successive evals didn't\
18
  \ cover the same tasks. You find each in the results and the \"latest\" split for\
19
+ \ each eval):\n\n```python\n{\n \"all\": {\n \"em\": 0.037751677852348994,\n\
20
+ \ \"em_stderr\": 0.0019518721243716466,\n \"f1\": 0.10559354026845587,\n\
21
+ \ \"f1_stderr\": 0.00235422441511938,\n \"acc\": 0.4481872912020479,\n\
22
+ \ \"acc_stderr\": 0.010817015780168433\n },\n \"harness|drop|3\": {\n\
23
+ \ \"em\": 0.037751677852348994,\n \"em_stderr\": 0.0019518721243716466,\n\
24
+ \ \"f1\": 0.10559354026845587,\n \"f1_stderr\": 0.00235422441511938\n\
25
+ \ },\n \"harness|gsm8k|5\": {\n \"acc\": 0.14025777103866566,\n \
26
+ \ \"acc_stderr\": 0.009565108281428673\n },\n \"harness|winogrande|5\"\
27
+ : {\n \"acc\": 0.7561168113654302,\n \"acc_stderr\": 0.012068923278908192\n\
28
+ \ }\n}\n```"
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
29
  repo_url: https://huggingface.co/Undi95/ReMM-v2.2-L2-13B
30
  leaderboard_url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard
31
  point_of_contact: clementine@hf.co
 
38
  - split: latest
39
  path:
40
  - '**/details_harness|arc:challenge|25_2023-10-03T16-45-21.105610.parquet'
41
+ - config_name: harness_drop_3
42
+ data_files:
43
+ - split: 2023_10_23T13_54_57.235808
44
+ path:
45
+ - '**/details_harness|drop|3_2023-10-23T13-54-57.235808.parquet'
46
+ - split: latest
47
+ path:
48
+ - '**/details_harness|drop|3_2023-10-23T13-54-57.235808.parquet'
49
+ - config_name: harness_gsm8k_5
50
+ data_files:
51
+ - split: 2023_10_23T13_54_57.235808
52
+ path:
53
+ - '**/details_harness|gsm8k|5_2023-10-23T13-54-57.235808.parquet'
54
+ - split: latest
55
+ path:
56
+ - '**/details_harness|gsm8k|5_2023-10-23T13-54-57.235808.parquet'
57
  - config_name: harness_hellaswag_10
58
  data_files:
59
  - split: 2023_10_03T16_45_21.105610
 
646
  - split: latest
647
  path:
648
  - '**/details_harness|truthfulqa:mc|0_2023-10-03T16-45-21.105610.parquet'
649
+ - config_name: harness_winogrande_5
650
+ data_files:
651
+ - split: 2023_10_23T13_54_57.235808
652
+ path:
653
+ - '**/details_harness|winogrande|5_2023-10-23T13-54-57.235808.parquet'
654
+ - split: latest
655
+ path:
656
+ - '**/details_harness|winogrande|5_2023-10-23T13-54-57.235808.parquet'
657
  - config_name: results
658
  data_files:
659
  - split: 2023_10_03T16_45_21.105610
660
  path:
661
  - results_2023-10-03T16-45-21.105610.parquet
662
+ - split: 2023_10_23T13_54_57.235808
663
+ path:
664
+ - results_2023-10-23T13-54-57.235808.parquet
665
  - split: latest
666
  path:
667
+ - results_2023-10-23T13-54-57.235808.parquet
668
  ---
669
 
670
  # Dataset Card for Evaluation run of Undi95/ReMM-v2.2-L2-13B
 
681
 
682
  Dataset automatically created during the evaluation run of model [Undi95/ReMM-v2.2-L2-13B](https://huggingface.co/Undi95/ReMM-v2.2-L2-13B) on the [Open LLM Leaderboard](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard).
683
 
684
+ The dataset is composed of 64 configuration, each one coresponding to one of the evaluated task.
685
 
686
+ The dataset has been created from 2 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.
687
 
688
  An additional configuration "results" store all the aggregated results of the run (and is used to compute and display the agregated metrics on the [Open LLM Leaderboard](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)).
689
 
 
691
  ```python
692
  from datasets import load_dataset
693
  data = load_dataset("open-llm-leaderboard/details_Undi95__ReMM-v2.2-L2-13B",
694
+ "harness_winogrande_5",
695
  split="train")
696
  ```
697
 
698
  ## Latest results
699
 
700
+ These are the [latest results from run 2023-10-23T13:54:57.235808](https://huggingface.co/datasets/open-llm-leaderboard/details_Undi95__ReMM-v2.2-L2-13B/blob/main/results_2023-10-23T13-54-57.235808.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):
701
 
702
  ```python
703
  {
704
  "all": {
705
+ "em": 0.037751677852348994,
706
+ "em_stderr": 0.0019518721243716466,
707
+ "f1": 0.10559354026845587,
708
+ "f1_stderr": 0.00235422441511938,
709
+ "acc": 0.4481872912020479,
710
+ "acc_stderr": 0.010817015780168433
711
+ },
712
+ "harness|drop|3": {
713
+ "em": 0.037751677852348994,
714
+ "em_stderr": 0.0019518721243716466,
715
+ "f1": 0.10559354026845587,
716
+ "f1_stderr": 0.00235422441511938
717
+ },
718
+ "harness|gsm8k|5": {
719
+ "acc": 0.14025777103866566,
720
+ "acc_stderr": 0.009565108281428673
721
+ },
722
+ "harness|winogrande|5": {
723
+ "acc": 0.7561168113654302,
724
+ "acc_stderr": 0.012068923278908192
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
725
  }
726
  }
727
  ```