H2H-eval-comparator

Sleeping

rohansampath commited on Feb 12

Commit

8789e45

verified ·

1 Parent(s): 77d4add

Update mmlu_eval

Files changed (1) hide show

mmlu_eval CHANGED Viewed

@@ -8,7 +8,7 @@ from transformers import AutoTokenizer, AutoModelForCausalLM
 accuracy_metric = evaluate.load("accuracy")
 # Load MMLU dataset
-mmlu_dataset = load_dataset("lukaemon/mmlu")
 def generate_answer(model, tokenizer, question):
     """

 accuracy_metric = evaluate.load("accuracy")
 # Load MMLU dataset
+mmlu_dataset = load_dataset("cais/mmlu", "all")
 def generate_answer(model, tokenizer, question):
     """