Merge branch 'mmlu-pro' of github.com:ysjprojects/lm-evaluation-harness into mmlu-pro

Resolve conflict.

lm_eval/tasks/squad_completion/task.py

-"""
-"""
 import re
 from typing import List
 

lm_eval/tasks/squadv2/task.py

@@ -13,6 +13,7 @@ also determine when no answer is supported by the paragraph and abstain from ans
 
 Homepage: https://rajpurkar.github.io/SQuAD-explorer/
 """
+
 from functools import partial
 from math import exp
 

lm_eval/tasks/tinyBenchmarks/utils_winogrande.py

-""" This code mirrors the utils of the original winogrande task """
+"""This code mirrors the utils of the original winogrande task"""
 
 
 def doc_to_text(doc):

lm_eval/tasks/tmmluplus/default/_generate_configs.py

 """
 Take in a YAML, and output all "other" splits with this YAML
 """
+
 import argparse
 import os
 

scripts/clean_training_data/README.md

@@ -10,7 +10,7 @@ It uses the approach described in the [GPT-3 paper](https://arxiv.org/abs/2005.1
     the match, splitting the training data into chunks
    3) Any chunks less than `minimum_slice_length` are removed
    4) Training data sets split into more than `too_dirty_cutoff` are considered
-    completey contaminated and removed
+    completely contaminated and removed
 
 OpenAI used:
 ```

scripts/make_table_results.py

@@ -2,6 +2,7 @@
 Usage:
    python make_table_tasks.py --output <markdown_filename>
 """
+
 import json
 import logging
 import os

scripts/make_table_tasks.py

@@ -2,6 +2,7 @@
 Usage:
    python make_table_tasks.py --output <markdown_filename>
 """
+
 import argparse
 import logging