{"id":27438353,"url":"https://github.com/coderpat/croissant-llm-training","last_synced_at":"2025-04-14T20:35:19.040Z","repository":{"id":220515056,"uuid":"751430515","full_name":"CoderPat/croissant-llm-training","owner":"CoderPat","description":"Repository containing the code for training the CroissantLLM","archived":false,"fork":false,"pushed_at":"2024-02-02T13:00:16.000Z","size":15,"stargazers_count":0,"open_issues_count":0,"forks_count":0,"subscribers_count":1,"default_branch":"main","last_synced_at":"2024-02-02T14:24:41.453Z","etag":null,"topics":[],"latest_commit_sha":null,"homepage":null,"language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":null,"status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/CoderPat.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":null,"code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null}},"created_at":"2024-02-01T15:45:34.000Z","updated_at":"2024-02-02T14:24:43.172Z","dependencies_parsed_at":"2024-02-02T14:34:45.003Z","dependency_job_id":null,"html_url":"https://github.com/CoderPat/croissant-llm-training","commit_stats":null,"previous_names":["coderpat/croissant-llm-training"],"tags_count":0,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/CoderPat%2Fcroissant-llm-training","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/CoderPat%2Fcroissant-llm-training/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/CoderPat%2Fcroissant-llm-training/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/CoderPat%2Fcroissant-llm-training/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/CoderPat","download_url":"https://codeload.github.com/CoderPat/croissant-llm-training/tar.gz/refs/heads/main","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":248956974,"owners_count":21189460,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2022-07-04T15:15:14.044Z","host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":[],"created_at":"2025-04-14T20:35:18.456Z","updated_at":"2025-04-14T20:35:19.012Z","avatar_url":"https://github.com/CoderPat.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"# CroissantLLM: Training Repository\n\n## Installation Instructions\n\nAs a pre-requisite, make sure you have [ducttape](https://github.com/CoderPat/ducttape) and [(mini)conda](https://docs.conda.io/en/latest/miniconda.html) installed.\n\nFirst, clone this repository and its submodules:\n\n```bash\ngit clone --recurse-submodules git@github.com:CoderPat/croissant-llm-training.git\n```\n\nThen, to create a new conda environment with all the necessary dependencies, run the following command:\n\n```bash\nexport CONDA_HOME=\"/path/to/(mini)conda3\"\nbash setup/conda.sh\n```\n\n## Running pipelines\n\nThe core experimentation and training pipelines rely on ducttape, and are defined in `main.tape`. \nConfiguration files for different models and datasets are defined in `configs/`.\n\nStart by creating a configuration with user-dependent variables (like the output folder) in associated `configs/*_uservars.conf` associated with your chosen `.tconf`. E.g, for the `configs/croissant_llm.tconf` configuration, create a `configs/croissant_llm_uservars.conf` file with the following content:\n```\nglobal {\n    ducttape_output=/path/to/output\n    repo=/path/to/croissant-llm-training\n\n    (...)\n    # use a simple shell submitter \n    # we are forced to explicitly set the submitter parameters\n    # to make it compatible with other submitters (ie the slurm submitter)\n    submitter=shell\n    dump_account=none\n    dump_partition=none\n    (...)\n}\n```\n\nWe provide a template for our user variables used in JeanZay.\n\nThen, you can ran the one of the specified pipelines in `main.tape` by running ducttape with the corresponding configuration file:\n\n```bash\nconda activate towerllm-env\nducttape main.tape -C configs/croissant_llm.conf \n```","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fcoderpat%2Fcroissant-llm-training","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fcoderpat%2Fcroissant-llm-training","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fcoderpat%2Fcroissant-llm-training/lists"}