hendrycks/test issues and pull requests

#26 - Incorrect answers

Issue - State: open - Opened by cinjon about 2 months ago

#25 - What are expected to submit for the leaderboard integration?

Issue - State: open - Opened by zhimin-z 8 months ago

#24 - Suggestion: appropriate use of medical word: catheter

Issue - State: open - Opened by hiroyaiizuka 9 months ago

#23 - medical word typo in CSV file.

Issue - State: closed - Opened by hiroyaiizuka 9 months ago - 1 comment

#22 - Why is this repository called `test`?

Issue - State: open - Opened by nikhilweee 10 months ago

#21 - Duplicate Answers in Validation Set

Issue - State: open - Opened by riedgar-ms 10 months ago

#20 - can not download

Issue - State: open - Opened by fourfireM 10 months ago - 1 comment

#19 - Hrvatski

Issue - State: open - Opened by Deni7s 11 months ago

#18 - Incorrect answer for Q5 in high_school_computer_science_dev.csv

Issue - State: open - Opened by TenType 11 months ago

#17 - About mmlu subset "all" from hugging face

Issue - State: closed - Opened by githubhyz about 1 year ago

#16 - Issues with the moral scenarios task

Issue - State: closed - Opened by c1505 about 1 year ago - 3 comments

#15 - why ["top_logprobs"][-1]

Issue - State: closed - Opened by don-tpanic about 1 year ago - 1 comment

#14 - Seems that setting logprobs=100 is not useful now.

Issue - State: open - Opened by KL4805 about 1 year ago - 1 comment

#13 - Evaluation script for Huggingface Causal models

Pull Request - State: open - Opened by ollmer over 1 year ago - 15 comments

#12 - Update crop.py

Pull Request - State: closed - Opened by sbmaruf over 1 year ago

#11 - Answers A, B, C, D are not all equally likely - is it really accurate to use random baseline as comparison?

Issue - State: open - Opened by bmosaicml over 1 year ago

#10 - Dismatch dataset categories

Issue - State: open - Opened by zhichengg over 1 year ago

#9 - Fetching encoder json and bpe does not work (fixed by removing a typo)

Issue - State: open - Opened by kovacgrgur over 1 year ago

#8 - Flan-T5 Benchmarking Code

Pull Request - State: closed - Opened by Helw150 almost 2 years ago

#7 - Dataset size mismatched with paper

Issue - State: closed - Opened by todpole3 about 2 years ago - 2 comments

#6 - Unintended (?) repetition in moral_scenarios_val.csv

Issue - State: closed - Opened by sleepinyourhat about 2 years ago - 2 comments

#5 - Odd-Looking Samples in business_ethics_val.csv

Issue - State: closed - Opened by henighan over 2 years ago - 2 comments

#4 - Update README.md with papers with code mirror

Pull Request - State: open - Opened by RJT1990 over 2 years ago

#3 - Human level performance?

Issue - State: closed - Opened by rodrigonogueira4 about 4 years ago - 5 comments

#2 - Allows multiple engine selection from command line

Pull Request - State: closed - Opened by xksteven about 4 years ago

#1 - Adds downloading to make code more portable