Merge branch 'big-refactor' of...

Merge branch 'big-refactor' of https://github.com/EleutherAI/lm-evaluation-harness into port_master_squadv2

lm_eval/tasks/mmlu/flan_n_shot/loglikelihood/mmlu_us_foreign_policy.yaml

 "dataset_name": "us_foreign_policy"
 "description": "The following are multiple choice questions (with answers) about us\
   \ foreign policy.\n\n"
-"group": "mmlu_flan_n_shot_generative_social_sciences"
+"group": "mmlu_flan_n_shot_loglikelihood_social_sciences"
 "include": "_mmlu_flan_loglikelihood_template_yaml"
-"task": "mmlu_flan_n_shot_generative_us_foreign_policy"
+"task": "mmlu_flan_n_shot_loglikelihood_us_foreign_policy"

lm_eval/tasks/mmlu/flan_n_shot/loglikelihood/mmlu_virology.yaml

 "dataset_name": "virology"
 "description": "The following are multiple choice questions (with answers) about virology.\n\
   \n"
-"group": "mmlu_flan_n_shot_generative_other"
+"group": "mmlu_flan_n_shot_loglikelihood_other"
 "include": "_mmlu_flan_loglikelihood_template_yaml"
-"task": "mmlu_flan_n_shot_generative_virology"
+"task": "mmlu_flan_n_shot_loglikelihood_virology"

lm_eval/tasks/mmlu/flan_n_shot/loglikelihood/mmlu_world_religions.yaml

 "dataset_name": "world_religions"
 "description": "The following are multiple choice questions (with answers) about world\
   \ religions.\n\n"
-"group": "mmlu_flan_n_shot_generative_humanities"
+"group": "mmlu_flan_n_shot_loglikelihood_humanities"
 "include": "_mmlu_flan_loglikelihood_template_yaml"
-"task": "mmlu_flan_n_shot_generative_world_religions"
+"task": "mmlu_flan_n_shot_loglikelihood_world_religions"

lm_eval/utils.py

@@ -305,6 +305,10 @@ def make_table(result_dict, column: str = "results"):
 
     for k, dic in result_dict[column].items():
         version = result_dict["versions"][k]
+
+        if "alias" in dic:
+            k = dic.pop("alias")
+
         for (mf), v in dic.items():
             m, _, f = mf.partition(",")
             if m.endswith("_stderr"):