{"adoption": {"forks": 673, "observed_at": "2026-08-28T04:10:08.365950+00:00", "stars": 7868}, "canonical_url": "https://ross.abutalabs.com/products/palm-rlhf-pytorch", "card": {"archived": false, "artifact_type": "library", "description": "Implementation of RLHF (Reinforcement Learning with Human Feedback) on top of the PaLM architecture. Basically ChatGPT but with PaLM", "domain": ["large-language-models", "reinforcement-learning", "deep-learning", "artificial-intelligence"], "enriched": true, "function": ["llm-training", "reinforcement-learning", "machine-learning", "deep-learning"], "health_score": 92, "homepage": null, "language": "Python", "license": "MIT", "license_family": "permissive", "maturity": "experimental", "member_repos": ["lucidrains/PaLM-rlhf-pytorch"], "name": "lucidrains/PaLM-rlhf-pytorch", "platform": ["python"], "pushed_at": "2026-07-27T23:39:41+00:00", "repo": "lucidrains/PaLM-rlhf-pytorch", "stars": 7868, "tags": ["rlhf", "palm", "reward-model", "ppo", "transformers", "research-code", "open-source-chatgpt"], "topics": ["artificial-intelligence", "attention-mechanisms", "deep-learning", "reinforcement-learning", "transformers", "human-feedback"], "urls": [], "use_cases": ["train a language model with reinforcement learning from human feedback", "implement RLHF on a PaLM-style transformer in pytorch", "build an open-source ChatGPT-like model from scratch", "study how reward models and PPO are implemented for LLMs", "experiment with human preference fine-tuning of transformers"], "what_it_is": "A PyTorch library implementing Reinforcement Learning from Human Feedback (RLHF) on top of the PaLM transformer architecture, aiming to replicate ChatGPT-style training in the open. It provides the training scaffolding (reward model, PPO) but ships no pretrained model weights.", "when_to_avoid": ["you need a ready-to-use chatbot or pretrained model for inference", "you want a production-grade, well-supported RLHF framework", "you prefer simpler preference optimization methods like DPO over PPO-based RLHF"], "when_to_choose": ["you want readable reference code for RLHF components like reward modeling and PPO", "you are doing research on preference-based fine-tuning and want a customizable starting point", "you already have compute and data and need the RLHF training scaffolding"]}, "data_as_of": "2026-08-30T08:39:29.467469+00:00", "members": [{"path": "/products/palm-rlhf-pytorch", "repo": "lucidrains/PaLM-rlhf-pytorch", "role": "main", "score": 78}], "provenance": {"archived": {"kind": "observed", "observed_at": "2026-08-28T04:10:08.365950+00:00", "source": "github"}, "artifact_type": {"confidence": null, "enriched_at": "2026-08-29T17:33:45.145677+00:00", "kind": "inferred", "prompt_version": 1, "sources": [{"content_hash": "2256e18630462ffb6162656643ebfb8a6224308e864c37588999dfa24776402e", "fetched_at": "2026-08-28T04:10:08.365950+00:00", "kind": "readme", "missing": false, "url": "https://github.com/lucidrains/PaLM-rlhf-pytorch"}], "taxonomy_version": 1}, "description": {"kind": "observed", "observed_at": "2026-08-28T04:10:08.365950+00:00", "source": "github"}, "domain": {"confidence": null, "enriched_at": "2026-08-29T17:33:45.145677+00:00", "kind": "inferred", "prompt_version": 1, "sources": [{"content_hash": "2256e18630462ffb6162656643ebfb8a6224308e864c37588999dfa24776402e", "fetched_at": "2026-08-28T04:10:08.365950+00:00", "kind": "readme", "missing": false, "url": "https://github.com/lucidrains/PaLM-rlhf-pytorch"}], "taxonomy_version": 1}, "enriched": {"inputs": [], "kind": "computed", "method": "enrichment_status"}, "function": {"confidence": null, "enriched_at": "2026-08-29T17:33:45.145677+00:00", "kind": "inferred", "prompt_version": 1, "sources": [{"content_hash": "2256e18630462ffb6162656643ebfb8a6224308e864c37588999dfa24776402e", "fetched_at": "2026-08-28T04:10:08.365950+00:00", "kind": "readme", "missing": false, "url": "https://github.com/lucidrains/PaLM-rlhf-pytorch"}], "taxonomy_version": 1}, "health_score": {"inputs": ["days_since_push", "days_since_release", "archived"], "kind": "computed", "method": "health_v1"}, "homepage": {"kind": "observed", "observed_at": "2026-08-28T04:10:08.365950+00:00", "source": "github"}, "language": {"kind": "observed", "observed_at": "2026-08-28T04:10:08.365950+00:00", "source": "github"}, "license": {"kind": "observed", "observed_at": "2026-08-28T04:10:08.365950+00:00", "source": "github"}, "license_family": {"inputs": ["license"], "kind": "computed", "method": "license_family"}, "maturity": {"confidence": null, "enriched_at": "2026-08-29T17:33:45.145677+00:00", "kind": "inferred", "prompt_version": 1, "sources": [{"content_hash": "2256e18630462ffb6162656643ebfb8a6224308e864c37588999dfa24776402e", "fetched_at": "2026-08-28T04:10:08.365950+00:00", "kind": "readme", "missing": false, "url": "https://github.com/lucidrains/PaLM-rlhf-pytorch"}], "taxonomy_version": 1}, "member_repos": {"kind": "observed", "observed_at": "2026-08-28T04:10:08.365950+00:00", "source": "github"}, "name": {"kind": "observed", "observed_at": "2026-08-28T04:10:08.365950+00:00", "source": "github"}, "platform": {"confidence": null, "enriched_at": "2026-08-29T17:33:45.145677+00:00", "kind": "inferred", "prompt_version": 1, "sources": [{"content_hash": "2256e18630462ffb6162656643ebfb8a6224308e864c37588999dfa24776402e", "fetched_at": "2026-08-28T04:10:08.365950+00:00", "kind": "readme", "missing": false, "url": "https://github.com/lucidrains/PaLM-rlhf-pytorch"}], "taxonomy_version": 1}, "pushed_at": {"kind": "observed", "observed_at": "2026-08-28T04:10:08.365950+00:00", "source": "github"}, "repo": {"kind": "observed", "observed_at": "2026-08-28T04:10:08.365950+00:00", "source": "github"}, "stars": {"kind": "observed", "observed_at": "2026-08-28T04:10:08.365950+00:00", "source": "github"}, "tags": {"confidence": null, "enriched_at": "2026-08-29T17:33:45.145677+00:00", "kind": "inferred", "prompt_version": 1, "sources": [{"content_hash": "2256e18630462ffb6162656643ebfb8a6224308e864c37588999dfa24776402e", "fetched_at": "2026-08-28T04:10:08.365950+00:00", "kind": "readme", "missing": false, "url": "https://github.com/lucidrains/PaLM-rlhf-pytorch"}], "taxonomy_version": 1}, "topics": {"kind": "observed", "observed_at": "2026-08-28T04:10:08.365950+00:00", "source": "github"}, "urls": {"kind": "observed", "observed_at": "2026-08-28T04:10:08.365950+00:00", "source": "github"}, "use_cases": {"confidence": null, "enriched_at": "2026-08-29T17:33:45.145677+00:00", "kind": "inferred", "prompt_version": 1, "sources": [{"content_hash": "2256e18630462ffb6162656643ebfb8a6224308e864c37588999dfa24776402e", "fetched_at": "2026-08-28T04:10:08.365950+00:00", "kind": "readme", "missing": false, "url": "https://github.com/lucidrains/PaLM-rlhf-pytorch"}], "taxonomy_version": 1}, "what_it_is": {"confidence": null, "enriched_at": "2026-08-29T17:33:45.145677+00:00", "kind": "inferred", "prompt_version": 1, "sources": [{"content_hash": "2256e18630462ffb6162656643ebfb8a6224308e864c37588999dfa24776402e", "fetched_at": "2026-08-28T04:10:08.365950+00:00", "kind": "readme", "missing": false, "url": "https://github.com/lucidrains/PaLM-rlhf-pytorch"}], "taxonomy_version": 1}, "when_to_avoid": {"confidence": null, "enriched_at": "2026-08-29T17:33:45.145677+00:00", "kind": "inferred", "prompt_version": 1, "sources": [{"content_hash": "2256e18630462ffb6162656643ebfb8a6224308e864c37588999dfa24776402e", "fetched_at": "2026-08-28T04:10:08.365950+00:00", "kind": "readme", "missing": false, "url": "https://github.com/lucidrains/PaLM-rlhf-pytorch"}], "taxonomy_version": 1}, "when_to_choose": {"confidence": null, "enriched_at": "2026-08-29T17:33:45.145677+00:00", "kind": "inferred", "prompt_version": 1, "sources": [{"content_hash": "2256e18630462ffb6162656643ebfb8a6224308e864c37588999dfa24776402e", "fetched_at": "2026-08-28T04:10:08.365950+00:00", "kind": "readme", "missing": false, "url": "https://github.com/lucidrains/PaLM-rlhf-pytorch"}], "taxonomy_version": 1}}, "score": {"components": {"activity": 94, "longevity": 97, "rhythm": 48}, "computed_at": "2026-09-03T02:20:16.233290+00:00", "flags": [], "formula": "round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)", "inputs": {"age_days": 1363, "days_push": 37, "days_rel": 348, "gap_med": 0.0, "n_releases_24m": 21}, "score": 78, "version": 2}, "staleness": {"enrichment_outdated": false, "low_confidence": false, "scrape_days": 9, "stale_scrape": false}}