├── CLAUDE.md ├── EducationalGRPO.ipynb ├── OneShotAquaRAT.ipynb ├── PublicWorkingGRPO copy.ipynb ├── README2.md ├── StoneyGRPO.ipynb ├── configs └── math_grpo.yaml ├── docs ├── GOOGLE_CLOUD_STORAGE.md ├── GRADIO_DEPLOYMENT.md ├── PRIME_INTELLECT.md ├── README.md ├── WANDB_VISUALIZATION.md └── math_grpo.md ├── hf_grpotuned_pipeline.py ├── inspect_grpo_signature.py ├── pyproject.toml ├── src ├── oneshot_grpo.egg-info │ ├── PKG-INFO │ ├── SOURCES.txt │ ├── dependency_links.txt │ ├── requires.txt │ └── top_level.txt └── oneshot_grpo │ ├── __init__.py │ ├── __pycache__ │ └── __init__.cpython-312.pyc │ ├── data │ ├── __init__.py │ ├── __pycache__ │ │ ├── __init__.cpython-312.pyc │ │ └── gsm8k.cpython-312.pyc │ └── gsm8k.py │ ├── eval │ └── __init__.py │ ├── inference │ ├── __init__.py │ ├── __pycache__ │ │ ├── __init__.cpython-312.pyc │ │ └── pipeline.cpython-312.pyc │ └── pipeline.py │ ├── rewards │ ├── __init__.py │ ├── __pycache__ │ │ ├── __init__.cpython-312.pyc │ │ └── math.cpython-312.pyc │ └── math.py │ └── train │ ├── __init__.py │ ├── __pycache__ │ ├── __init__.cpython-312.pyc │ ├── grpo.cpython-312.pyc │ ├── math_pipeline.cpython-312.pyc │ └── run_math_grpo.cpython-312.pyc │ ├── grpo.py │ ├── math_pipeline.py │ └── run_math_grpo.py └── wandb └── run-20251009_143907-1vvw5i7h ├── files ├── output.log ├── requirements.txt └── wandb-metadata.json ├── logs ├── debug-core.log ├── debug-internal.log └── debug.log └── run-1vvw5i7h.wandb /CLAUDE.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/HarleyCoops/OneShotAquaRAT/HEAD/CLAUDE.md -------------------------------------------------------------------------------- /EducationalGRPO.ipynb: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /OneShotAquaRAT.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/HarleyCoops/OneShotAquaRAT/HEAD/OneShotAquaRAT.ipynb -------------------------------------------------------------------------------- /PublicWorkingGRPO copy.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/HarleyCoops/OneShotAquaRAT/HEAD/PublicWorkingGRPO copy.ipynb -------------------------------------------------------------------------------- /README2.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/HarleyCoops/OneShotAquaRAT/HEAD/README2.md -------------------------------------------------------------------------------- /StoneyGRPO.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/HarleyCoops/OneShotAquaRAT/HEAD/StoneyGRPO.ipynb -------------------------------------------------------------------------------- /configs/math_grpo.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/HarleyCoops/OneShotAquaRAT/HEAD/configs/math_grpo.yaml -------------------------------------------------------------------------------- /docs/GOOGLE_CLOUD_STORAGE.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/HarleyCoops/OneShotAquaRAT/HEAD/docs/GOOGLE_CLOUD_STORAGE.md -------------------------------------------------------------------------------- /docs/GRADIO_DEPLOYMENT.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/HarleyCoops/OneShotAquaRAT/HEAD/docs/GRADIO_DEPLOYMENT.md -------------------------------------------------------------------------------- /docs/PRIME_INTELLECT.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/HarleyCoops/OneShotAquaRAT/HEAD/docs/PRIME_INTELLECT.md -------------------------------------------------------------------------------- /docs/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/HarleyCoops/OneShotAquaRAT/HEAD/docs/README.md -------------------------------------------------------------------------------- /docs/WANDB_VISUALIZATION.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/HarleyCoops/OneShotAquaRAT/HEAD/docs/WANDB_VISUALIZATION.md -------------------------------------------------------------------------------- /docs/math_grpo.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/HarleyCoops/OneShotAquaRAT/HEAD/docs/math_grpo.md -------------------------------------------------------------------------------- /hf_grpotuned_pipeline.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/HarleyCoops/OneShotAquaRAT/HEAD/hf_grpotuned_pipeline.py -------------------------------------------------------------------------------- /inspect_grpo_signature.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/HarleyCoops/OneShotAquaRAT/HEAD/inspect_grpo_signature.py -------------------------------------------------------------------------------- /pyproject.toml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/HarleyCoops/OneShotAquaRAT/HEAD/pyproject.toml -------------------------------------------------------------------------------- /src/oneshot_grpo.egg-info/PKG-INFO: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/HarleyCoops/OneShotAquaRAT/HEAD/src/oneshot_grpo.egg-info/PKG-INFO -------------------------------------------------------------------------------- /src/oneshot_grpo.egg-info/SOURCES.txt: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/HarleyCoops/OneShotAquaRAT/HEAD/src/oneshot_grpo.egg-info/SOURCES.txt -------------------------------------------------------------------------------- /src/oneshot_grpo.egg-info/dependency_links.txt: -------------------------------------------------------------------------------- 1 | 2 | -------------------------------------------------------------------------------- /src/oneshot_grpo.egg-info/requires.txt: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/HarleyCoops/OneShotAquaRAT/HEAD/src/oneshot_grpo.egg-info/requires.txt -------------------------------------------------------------------------------- /src/oneshot_grpo.egg-info/top_level.txt: -------------------------------------------------------------------------------- 1 | oneshot_grpo 2 | -------------------------------------------------------------------------------- /src/oneshot_grpo/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/HarleyCoops/OneShotAquaRAT/HEAD/src/oneshot_grpo/__init__.py -------------------------------------------------------------------------------- /src/oneshot_grpo/__pycache__/__init__.cpython-312.pyc: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/HarleyCoops/OneShotAquaRAT/HEAD/src/oneshot_grpo/__pycache__/__init__.cpython-312.pyc -------------------------------------------------------------------------------- /src/oneshot_grpo/data/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/HarleyCoops/OneShotAquaRAT/HEAD/src/oneshot_grpo/data/__init__.py -------------------------------------------------------------------------------- /src/oneshot_grpo/data/__pycache__/__init__.cpython-312.pyc: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/HarleyCoops/OneShotAquaRAT/HEAD/src/oneshot_grpo/data/__pycache__/__init__.cpython-312.pyc -------------------------------------------------------------------------------- /src/oneshot_grpo/data/__pycache__/gsm8k.cpython-312.pyc: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/HarleyCoops/OneShotAquaRAT/HEAD/src/oneshot_grpo/data/__pycache__/gsm8k.cpython-312.pyc -------------------------------------------------------------------------------- /src/oneshot_grpo/data/gsm8k.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/HarleyCoops/OneShotAquaRAT/HEAD/src/oneshot_grpo/data/gsm8k.py -------------------------------------------------------------------------------- /src/oneshot_grpo/eval/__init__.py: -------------------------------------------------------------------------------- 1 | """Evaluation utilities placeholder.""" 2 | -------------------------------------------------------------------------------- /src/oneshot_grpo/inference/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/HarleyCoops/OneShotAquaRAT/HEAD/src/oneshot_grpo/inference/__init__.py -------------------------------------------------------------------------------- /src/oneshot_grpo/inference/__pycache__/__init__.cpython-312.pyc: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/HarleyCoops/OneShotAquaRAT/HEAD/src/oneshot_grpo/inference/__pycache__/__init__.cpython-312.pyc -------------------------------------------------------------------------------- /src/oneshot_grpo/inference/__pycache__/pipeline.cpython-312.pyc: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/HarleyCoops/OneShotAquaRAT/HEAD/src/oneshot_grpo/inference/__pycache__/pipeline.cpython-312.pyc -------------------------------------------------------------------------------- /src/oneshot_grpo/inference/pipeline.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/HarleyCoops/OneShotAquaRAT/HEAD/src/oneshot_grpo/inference/pipeline.py -------------------------------------------------------------------------------- /src/oneshot_grpo/rewards/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/HarleyCoops/OneShotAquaRAT/HEAD/src/oneshot_grpo/rewards/__init__.py -------------------------------------------------------------------------------- /src/oneshot_grpo/rewards/__pycache__/__init__.cpython-312.pyc: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/HarleyCoops/OneShotAquaRAT/HEAD/src/oneshot_grpo/rewards/__pycache__/__init__.cpython-312.pyc -------------------------------------------------------------------------------- /src/oneshot_grpo/rewards/__pycache__/math.cpython-312.pyc: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/HarleyCoops/OneShotAquaRAT/HEAD/src/oneshot_grpo/rewards/__pycache__/math.cpython-312.pyc -------------------------------------------------------------------------------- /src/oneshot_grpo/rewards/math.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/HarleyCoops/OneShotAquaRAT/HEAD/src/oneshot_grpo/rewards/math.py -------------------------------------------------------------------------------- /src/oneshot_grpo/train/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/HarleyCoops/OneShotAquaRAT/HEAD/src/oneshot_grpo/train/__init__.py -------------------------------------------------------------------------------- /src/oneshot_grpo/train/__pycache__/__init__.cpython-312.pyc: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/HarleyCoops/OneShotAquaRAT/HEAD/src/oneshot_grpo/train/__pycache__/__init__.cpython-312.pyc -------------------------------------------------------------------------------- /src/oneshot_grpo/train/__pycache__/grpo.cpython-312.pyc: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/HarleyCoops/OneShotAquaRAT/HEAD/src/oneshot_grpo/train/__pycache__/grpo.cpython-312.pyc -------------------------------------------------------------------------------- /src/oneshot_grpo/train/__pycache__/math_pipeline.cpython-312.pyc: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/HarleyCoops/OneShotAquaRAT/HEAD/src/oneshot_grpo/train/__pycache__/math_pipeline.cpython-312.pyc -------------------------------------------------------------------------------- /src/oneshot_grpo/train/__pycache__/run_math_grpo.cpython-312.pyc: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/HarleyCoops/OneShotAquaRAT/HEAD/src/oneshot_grpo/train/__pycache__/run_math_grpo.cpython-312.pyc -------------------------------------------------------------------------------- /src/oneshot_grpo/train/grpo.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/HarleyCoops/OneShotAquaRAT/HEAD/src/oneshot_grpo/train/grpo.py -------------------------------------------------------------------------------- /src/oneshot_grpo/train/math_pipeline.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/HarleyCoops/OneShotAquaRAT/HEAD/src/oneshot_grpo/train/math_pipeline.py -------------------------------------------------------------------------------- /src/oneshot_grpo/train/run_math_grpo.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/HarleyCoops/OneShotAquaRAT/HEAD/src/oneshot_grpo/train/run_math_grpo.py -------------------------------------------------------------------------------- /wandb/run-20251009_143907-1vvw5i7h/files/output.log: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /wandb/run-20251009_143907-1vvw5i7h/files/requirements.txt: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/HarleyCoops/OneShotAquaRAT/HEAD/wandb/run-20251009_143907-1vvw5i7h/files/requirements.txt -------------------------------------------------------------------------------- /wandb/run-20251009_143907-1vvw5i7h/files/wandb-metadata.json: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/HarleyCoops/OneShotAquaRAT/HEAD/wandb/run-20251009_143907-1vvw5i7h/files/wandb-metadata.json -------------------------------------------------------------------------------- /wandb/run-20251009_143907-1vvw5i7h/logs/debug-core.log: -------------------------------------------------------------------------------- 1 | C:/Users/chris/AppData/Local/wandb/logs/core-debug-20251009_143908.log -------------------------------------------------------------------------------- /wandb/run-20251009_143907-1vvw5i7h/logs/debug-internal.log: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/HarleyCoops/OneShotAquaRAT/HEAD/wandb/run-20251009_143907-1vvw5i7h/logs/debug-internal.log -------------------------------------------------------------------------------- /wandb/run-20251009_143907-1vvw5i7h/logs/debug.log: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/HarleyCoops/OneShotAquaRAT/HEAD/wandb/run-20251009_143907-1vvw5i7h/logs/debug.log -------------------------------------------------------------------------------- /wandb/run-20251009_143907-1vvw5i7h/run-1vvw5i7h.wandb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/HarleyCoops/OneShotAquaRAT/HEAD/wandb/run-20251009_143907-1vvw5i7h/run-1vvw5i7h.wandb --------------------------------------------------------------------------------