{"adoption": {"forks": 133, "observed_at": "2026-08-28T04:05:10.965034+00:00", "stars": 1611}, "canonical_url": "https://ross.abutalabs.com/products/safe-rlhf", "card": {"archived": false, "artifact_type": "framework", "description": "Safe RLHF: Constrained Value Alignment via Safe Reinforcement Learning from Human Feedback", "domain": ["large-language-models", "machine-learning", "artificial-intelligence", "deep-learning"], "enriched": true, "function": ["llm-training", "machine-learning", "deep-learning", "rag"], "health_score": 52, "homepage": "https://pku-beaver.github.io", "language": "Python", "license": "Apache-2.0", "license_family": "permissive", "maturity": "active", "member_repos": ["PKU-Alignment/safe-rlhf"], "name": "PKU-Alignment/safe-rlhf", "platform": ["python"], "pushed_at": "2025-11-24T04:07:51+00:00", "repo": "PKU-Alignment/safe-rlhf", "stars": 1611, "tags": ["rlhf", "safe-rlhf", "alignment", "ai-safety", "reward-model", "cost-model", "sft", "deepspeed", "preference-learning", "llm-alignment", "gpu", "linux"], "topics": ["ai-safety", "alpaca", "datasets", "deepspeed", "large-language-models", "llama", "llm", "llms", "reinforcement-learning", "reinforcement-learning-from-human-feedback", "rlhf", "transformers", "vicuna", "safe-rlhf", "safe-reinforcement-learning", "safe-reinforcement-learning-from-human-feedback", "safety", "gpt", "transformer", "beaver"], "urls": [], "use_cases": ["train an LLM with RLHF", "align a language model with safety constraints", "run supervised fine-tuning on LLaMA or OPT", "train a reward model from human preferences", "train a cost model for harmlessness", "research safe reinforcement learning from human feedback", "evaluate LLM safety with BIG-bench or GPT-4 evaluation"], "what_it_is": "Beaver is a modular open-source framework from Peking University for training large language models with SFT, RLHF, and Safe RLHF (constrained reward maximization). It includes a large human-labeled preference dataset (up to 1M pairs) and pre-trained reward/cost model checkpoints.", "when_to_avoid": ["you only need to run inference on an existing LLM", "you lack multi-GPU resources for large model training", "you need a general-purpose RL library unrelated to LLM alignment"], "when_to_choose": ["you need a reproducible RLHF or Safe RLHF training pipeline", "you want human-labeled helpfulness and harmlessness preference data", "you are doing alignment research with reward and cost models", "you want pre-trained aligned model checkpoints like Beaver-7B"]}, "data_as_of": "2026-08-30T08:39:29.467469+00:00", "members": [{"path": "/products/safe-rlhf", "repo": "PKU-Alignment/safe-rlhf", "role": "main", "score": 53}], "provenance": {"archived": {"kind": "observed", "observed_at": "2026-08-28T04:05:10.965034+00:00", "source": "github"}, "artifact_type": {"confidence": null, "enriched_at": "2026-08-30T03:51:11.218352+00:00", "kind": "inferred", "prompt_version": 1, "sources": [{"content_hash": "0f1ee9303a3c4214ab5e690ae82e585929c196a207c77c38740ec585b3b4333b", "fetched_at": "2026-08-28T04:05:10.965034+00:00", "kind": "readme", "missing": false, "url": "https://github.com/PKU-Alignment/safe-rlhf"}, {"content_hash": "df9592ac827c475372599127fef0a5a2b2f0f009fb1441e0b1ce7536578c80ba", "fetched_at": "2026-08-29T11:23:08.713632+00:00", "kind": "homepage", "missing": false, "url": "https://pku-beaver.github.io"}], "taxonomy_version": 1}, "description": {"kind": "observed", "observed_at": "2026-08-28T04:05:10.965034+00:00", "source": "github"}, "domain": {"confidence": null, "enriched_at": "2026-08-30T03:51:11.218352+00:00", "kind": "inferred", "prompt_version": 1, "sources": [{"content_hash": "0f1ee9303a3c4214ab5e690ae82e585929c196a207c77c38740ec585b3b4333b", "fetched_at": "2026-08-28T04:05:10.965034+00:00", "kind": "readme", "missing": false, "url": "https://github.com/PKU-Alignment/safe-rlhf"}, {"content_hash": "df9592ac827c475372599127fef0a5a2b2f0f009fb1441e0b1ce7536578c80ba", "fetched_at": "2026-08-29T11:23:08.713632+00:00", "kind": "homepage", "missing": false, "url": "https://pku-beaver.github.io"}], "taxonomy_version": 1}, "enriched": {"inputs": [], "kind": "computed", "method": "enrichment_status"}, "function": {"confidence": null, "enriched_at": "2026-08-30T03:51:11.218352+00:00", "kind": "inferred", "prompt_version": 1, "sources": [{"content_hash": "0f1ee9303a3c4214ab5e690ae82e585929c196a207c77c38740ec585b3b4333b", "fetched_at": "2026-08-28T04:05:10.965034+00:00", "kind": "readme", "missing": false, "url": "https://github.com/PKU-Alignment/safe-rlhf"}, {"content_hash": "df9592ac827c475372599127fef0a5a2b2f0f009fb1441e0b1ce7536578c80ba", "fetched_at": "2026-08-29T11:23:08.713632+00:00", "kind": "homepage", "missing": false, "url": "https://pku-beaver.github.io"}], "taxonomy_version": 1}, "health_score": {"inputs": ["days_since_push", "days_since_release", "archived"], "kind": "computed", "method": "health_v1"}, "homepage": {"kind": "observed", "observed_at": "2026-08-28T04:05:10.965034+00:00", "source": "github"}, "language": {"kind": "observed", "observed_at": "2026-08-28T04:05:10.965034+00:00", "source": "github"}, "license": {"kind": "observed", "observed_at": "2026-08-28T04:05:10.965034+00:00", "source": "github"}, "license_family": {"inputs": ["license"], "kind": "computed", "method": "license_family"}, "maturity": {"confidence": null, "enriched_at": "2026-08-30T03:51:11.218352+00:00", "kind": "inferred", "prompt_version": 1, "sources": [{"content_hash": "0f1ee9303a3c4214ab5e690ae82e585929c196a207c77c38740ec585b3b4333b", "fetched_at": "2026-08-28T04:05:10.965034+00:00", "kind": "readme", "missing": false, "url": "https://github.com/PKU-Alignment/safe-rlhf"}, {"content_hash": "df9592ac827c475372599127fef0a5a2b2f0f009fb1441e0b1ce7536578c80ba", "fetched_at": "2026-08-29T11:23:08.713632+00:00", "kind": "homepage", "missing": false, "url": "https://pku-beaver.github.io"}], "taxonomy_version": 1}, "member_repos": {"kind": "observed", "observed_at": "2026-08-28T04:05:10.965034+00:00", "source": "github"}, "name": {"kind": "observed", "observed_at": "2026-08-28T04:05:10.965034+00:00", "source": "github"}, "platform": {"confidence": null, "enriched_at": "2026-08-30T03:51:11.218352+00:00", "kind": "inferred", "prompt_version": 1, "sources": [{"content_hash": "0f1ee9303a3c4214ab5e690ae82e585929c196a207c77c38740ec585b3b4333b", "fetched_at": "2026-08-28T04:05:10.965034+00:00", "kind": "readme", "missing": false, "url": "https://github.com/PKU-Alignment/safe-rlhf"}, {"content_hash": "df9592ac827c475372599127fef0a5a2b2f0f009fb1441e0b1ce7536578c80ba", "fetched_at": "2026-08-29T11:23:08.713632+00:00", "kind": "homepage", "missing": false, "url": "https://pku-beaver.github.io"}], "taxonomy_version": 1}, "pushed_at": {"kind": "observed", "observed_at": "2026-08-28T04:05:10.965034+00:00", "source": "github"}, "repo": {"kind": "observed", "observed_at": "2026-08-28T04:05:10.965034+00:00", "source": "github"}, "stars": {"kind": "observed", "observed_at": "2026-08-28T04:05:10.965034+00:00", "source": "github"}, "tags": {"confidence": null, "enriched_at": "2026-08-30T03:51:11.218352+00:00", "kind": "inferred", "prompt_version": 1, "sources": [{"content_hash": "0f1ee9303a3c4214ab5e690ae82e585929c196a207c77c38740ec585b3b4333b", "fetched_at": "2026-08-28T04:05:10.965034+00:00", "kind": "readme", "missing": false, "url": "https://github.com/PKU-Alignment/safe-rlhf"}, {"content_hash": "df9592ac827c475372599127fef0a5a2b2f0f009fb1441e0b1ce7536578c80ba", "fetched_at": "2026-08-29T11:23:08.713632+00:00", "kind": "homepage", "missing": false, "url": "https://pku-beaver.github.io"}], "taxonomy_version": 1}, "topics": {"kind": "observed", "observed_at": "2026-08-28T04:05:10.965034+00:00", "source": "github"}, "urls": {"kind": "observed", "observed_at": "2026-08-28T04:05:10.965034+00:00", "source": "github"}, "use_cases": {"confidence": null, "enriched_at": "2026-08-30T03:51:11.218352+00:00", "kind": "inferred", "prompt_version": 1, "sources": [{"content_hash": "0f1ee9303a3c4214ab5e690ae82e585929c196a207c77c38740ec585b3b4333b", "fetched_at": "2026-08-28T04:05:10.965034+00:00", "kind": "readme", "missing": false, "url": "https://github.com/PKU-Alignment/safe-rlhf"}, {"content_hash": "df9592ac827c475372599127fef0a5a2b2f0f009fb1441e0b1ce7536578c80ba", "fetched_at": "2026-08-29T11:23:08.713632+00:00", "kind": "homepage", "missing": false, "url": "https://pku-beaver.github.io"}], "taxonomy_version": 1}, "what_it_is": {"confidence": null, "enriched_at": "2026-08-30T03:51:11.218352+00:00", "kind": "inferred", "prompt_version": 1, "sources": [{"content_hash": "0f1ee9303a3c4214ab5e690ae82e585929c196a207c77c38740ec585b3b4333b", "fetched_at": "2026-08-28T04:05:10.965034+00:00", "kind": "readme", "missing": false, "url": "https://github.com/PKU-Alignment/safe-rlhf"}, {"content_hash": "df9592ac827c475372599127fef0a5a2b2f0f009fb1441e0b1ce7536578c80ba", "fetched_at": "2026-08-29T11:23:08.713632+00:00", "kind": "homepage", "missing": false, "url": "https://pku-beaver.github.io"}], "taxonomy_version": 1}, "when_to_avoid": {"confidence": null, "enriched_at": "2026-08-30T03:51:11.218352+00:00", "kind": "inferred", "prompt_version": 1, "sources": [{"content_hash": "0f1ee9303a3c4214ab5e690ae82e585929c196a207c77c38740ec585b3b4333b", "fetched_at": "2026-08-28T04:05:10.965034+00:00", "kind": "readme", "missing": false, "url": "https://github.com/PKU-Alignment/safe-rlhf"}, {"content_hash": "df9592ac827c475372599127fef0a5a2b2f0f009fb1441e0b1ce7536578c80ba", "fetched_at": "2026-08-29T11:23:08.713632+00:00", "kind": "homepage", "missing": false, "url": "https://pku-beaver.github.io"}], "taxonomy_version": 1}, "when_to_choose": {"confidence": null, "enriched_at": "2026-08-30T03:51:11.218352+00:00", "kind": "inferred", "prompt_version": 1, "sources": [{"content_hash": "0f1ee9303a3c4214ab5e690ae82e585929c196a207c77c38740ec585b3b4333b", "fetched_at": "2026-08-28T04:05:10.965034+00:00", "kind": "readme", "missing": false, "url": "https://github.com/PKU-Alignment/safe-rlhf"}, {"content_hash": "df9592ac827c475372599127fef0a5a2b2f0f009fb1441e0b1ce7536578c80ba", "fetched_at": "2026-08-29T11:23:08.713632+00:00", "kind": "homepage", "missing": false, "url": "https://pku-beaver.github.io"}], "taxonomy_version": 1}}, "score": {"components": {"activity": 53, "longevity": 86, "rhythm": 35}, "computed_at": "2026-09-02T17:46:02.011165+00:00", "flags": ["no_releases"], "formula": "round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)", "inputs": {"age_days": 1206, "days_push": 282, "days_rel": null, "gap_med": null, "n_releases_24m": 0}, "score": 53, "version": 2}, "staleness": {"enrichment_outdated": false, "low_confidence": false, "scrape_days": 9, "stale_scrape": false}}