npy and pkl not found in llm-topla\llm-topla-main\results\mmlu_hf extracted from "Basemodel Outputs" compressed file.

[xsthunder]

Hi, thanks for sharing the code. i got into trouble when trying the runing the mmlu_hf task.

as i dived into the code, i found pkl and npy is needed:

llm-topla/helper.py

Lines 158 to 165 in a01e555

	def load_mmlu_multi_run(model_names, dataset_name):
	data_dir = os.path.join(RESULT_DIR, "mmlu")
	# find number of samples for each model
	data = []
	choices = ["A", "B", "C", "D"]
	for model_n in model_names:
	arr_path = os.path.join(data_dir, model_n, dataset_name, f"latest_predictions.npy")
	outputs_path = os.path.join(data_dir, model_n, dataset_name, f"latest_outputs.pkl")

however pkl and npy is not presented in the compressed file ur provided. There is only csv files.

llm-topla/README.md

Lines 29 to 35 in a01e555

	## Basemodel Outputs
	The base model outputs of these datasets are can be downloaded from the link below and must be put inside `results/`
	```
	https://drive.google.com/drive/folders/17_5JN5koFKsnyty9klNz1nBGYLKSE4tw?usp=sharing
	```

PS: the gsm8k folder has npy and pkl and was successful