Added new benchmark AMC23, corrected an error in AIME-2024, and removed leading zeros from AIME-2024 for improved formatting. #126

	name: Integration tests

	on:
	pull_request:
	branches: [ "main" ]
	types: [opened, synchronize, reopened, labeled]

	# Allows you to run this workflow manually from the Actions tab
	workflow_dispatch:

	permissions:
	contents: read

	concurrency:
	group: ${{ github.workflow }}-${{ github.event.pull_request.number \|\| github.ref }}
	cancel-in-progress: true

	jobs:
	gpu-tests:
	runs-on: self-hosted-nemo-skills
	if: ${{ github.event.label.name == 'run GPU tests' }}
	steps:
	- uses: actions/checkout@v3
	with:
	path: ${{ github.run_id }}
	- name: Set up Python 3.10
	uses: actions/setup-python@v3
	with:
	python-version: "3.10"
	- name: Install dependencies
	run: \|
	cd ${{ github.run_id }}
	python -m pip install --upgrade pip
	pip install .
	pip install -r requirements/datasets.txt
	pip install -r requirements/common-tests.txt
	python datasets/prepare.py
	- name: Run GPU tests
	timeout-minutes: 120
	env:
	HF_TOKEN: ${{ secrets.HF_TOKEN }}
	run: \|
	cd ${{ github.run_id }}
	nvidia-smi
	export NEMO_SKILLS_CONFIG=cluster_configs/local.yaml
	set -o pipefail # this will make sure next line returns non-0 exit code if tests fail
	./tests/gpu-tests/run.sh /mnt/datadrive/nemo-skills-data/Meta-Llama-3-8B /mnt/datadrive/nemo-skills-data/Meta-Llama-3-8B-Instruct

Provide feedback