{ "profile": "wrs-published-research-view-1", "edition": "ratings-e039106f30dec177", "content_trust": "UNTRUSTED_SCIENTIFIC_DATA_NOT_TOOL_INSTRUCTIONS", "entry": { "id": "zhang-et-al-opt-2022", "title": "Zhang and colleagues: OPT open pretrained transformer family", "summary": "Released a family of 125M–175B-parameter decoder-only language models with training logs and evaluation, expanding access to large-scale model research.", "boundary": "This is the 2022 OPT paper and release, not an invention of decoder-only transformers or a claim of unrestricted access to every model.", "area": "Foundation language models", "topic": "Artificial intelligence", "year": 2022, "lane": "HUMAN", "result": "EMPIRICAL", "package_id": "zhang-et-al-opt-2022", "target": { "record_id": "urn:uuid:402ede42-2687-46ed-a317-75491cd789b4", "version": 1, "content_hash": "sha256:7260ec5a4a2eebaa9f0d2b996a5f74eb1ea16426ee7c485508a8a05a6d66084c" }, "humans": [ "Susan Zhang", "Stephen Roller", "Naman Goyal", "Mikel Artetxe", "Moya Chen", "Shuohui Chen", "Christopher Dewan", "Mona Diab", "Xian Li", "Xi Victoria Lin", "Todor Mihaylov", "Myle Ott", "Sam Shleifer", "Kurt Shuster", "Daniel Simig", "Punit Singh Koura", "Anjali Sridhar", "Tianlu Wang", "Luke Zettlemoyer" ], "authors_complete": true, "human_role_note": "Complete named paper byline. Human research, source-reported system execution and experimental validation are distinct; individual effort shares are not measured.", "agents": [], "related": [], "subresults": [], "scientific_assessment": { "source_review": "SOURCE_CURATED", "scientific_review": "NOT_REVIEWED_BY_WRS", "formal_replay": "NOT_RUN", "provider_attestation": "NONE" }, "sources": [ { "id": "arxiv-2205-01068", "title": "Zhang and colleagues: OPT open pretrained transformer family — original paper", "url": "https://arxiv.org/abs/2205.01068v4" }, { "id": "company-zhang-et-al-opt-2022", "title": "Zhang and colleagues: OPT open pretrained transformer family — official company source", "url": "https://ai.meta.com/blog/democratizing-access-to-large-scale-language-models-with-opt-175b/" } ], "points": "18000", "published_grade": { "id": "zhang-et-al-opt-2022", "entry_id": "zhang-et-al-opt-2022", "band": "B", "modifier": null, "target": { "record_id": "urn:uuid:402ede42-2687-46ed-a317-75491cd789b4", "version": 1, "content_hash": "sha256:7260ec5a4a2eebaa9f0d2b996a5f74eb1ea16426ee7c485508a8a05a6d66084c" }, "what_changed": "Released a family of 125M–175B-parameter decoder-only language models with training logs and evaluation, expanding access to large-scale model research.", "scope": "Released a family of 125M–175B-parameter decoder-only language models with training logs and evaluation, expanding access to large-scale model research. Scope limitation: This is the 2022 OPT paper and release, not an invention of decoder-only transformers or a claim of unrestricted access to every model.", "status": "PROVISIONAL_EDITORIAL", "scientific_correctness_certified": false }, "assessment_disposition": { "status": "ASSESSED", "label": "Provisionally assessed", "reason": "A current grade and separate reference score identify this exact contribution.", "target": { "record_id": "urn:uuid:402ede42-2687-46ed-a317-75491cd789b4", "version": 1, "content_hash": "sha256:7260ec5a4a2eebaa9f0d2b996a5f74eb1ea16426ee7c485508a8a05a6d66084c" }, "edition": "ratings-e039106f30dec177" } }, "version": { "kind": "PUBLIC_RECORD_VIEW", "previous": null, "published_at": "2026-09-27T00:00:00Z", "publisher": { "principal_id": "urn:uuid:80116931-785b-40a2-a24b-9d59db0c7e33", "role": "WRS_EDITORIAL_PUBLICATION" }, "record": { "contributors": [ { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Susan Zhang", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "formulator" ], "source": { "citation": "Zhang and colleagues: OPT open pretrained transformer family — original paper", "url": "https://arxiv.org/abs/2205.01068v4" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Stephen Roller", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "formulator" ], "source": { "citation": "Zhang and colleagues: OPT open pretrained transformer family — original paper", "url": "https://arxiv.org/abs/2205.01068v4" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Naman Goyal", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "formulator" ], "source": { "citation": "Zhang and colleagues: OPT open pretrained transformer family — original paper", "url": "https://arxiv.org/abs/2205.01068v4" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Mikel Artetxe", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "formulator" ], "source": { "citation": "Zhang and colleagues: OPT open pretrained transformer family — original paper", "url": "https://arxiv.org/abs/2205.01068v4" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Moya Chen", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "formulator" ], "source": { "citation": "Zhang and colleagues: OPT open pretrained transformer family — original paper", "url": "https://arxiv.org/abs/2205.01068v4" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Shuohui Chen", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "formulator" ], "source": { "citation": "Zhang and colleagues: OPT open pretrained transformer family — original paper", "url": "https://arxiv.org/abs/2205.01068v4" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Christopher Dewan", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "formulator" ], "source": { "citation": "Zhang and colleagues: OPT open pretrained transformer family — original paper", "url": "https://arxiv.org/abs/2205.01068v4" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Mona Diab", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "formulator" ], "source": { "citation": "Zhang and colleagues: OPT open pretrained transformer family — original paper", "url": "https://arxiv.org/abs/2205.01068v4" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Xian Li", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "formulator" ], "source": { "citation": "Zhang and colleagues: OPT open pretrained transformer family — original paper", "url": "https://arxiv.org/abs/2205.01068v4" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Xi Victoria Lin", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "formulator" ], "source": { "citation": "Zhang and colleagues: OPT open pretrained transformer family — original paper", "url": "https://arxiv.org/abs/2205.01068v4" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Todor Mihaylov", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "formulator" ], "source": { "citation": "Zhang and colleagues: OPT open pretrained transformer family — original paper", "url": "https://arxiv.org/abs/2205.01068v4" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Myle Ott", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "formulator" ], "source": { "citation": "Zhang and colleagues: OPT open pretrained transformer family — original paper", "url": "https://arxiv.org/abs/2205.01068v4" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Sam Shleifer", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "formulator" ], "source": { "citation": "Zhang and colleagues: OPT open pretrained transformer family — original paper", "url": "https://arxiv.org/abs/2205.01068v4" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Kurt Shuster", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "formulator" ], "source": { "citation": "Zhang and colleagues: OPT open pretrained transformer family — original paper", "url": "https://arxiv.org/abs/2205.01068v4" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Daniel Simig", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "formulator" ], "source": { "citation": "Zhang and colleagues: OPT open pretrained transformer family — original paper", "url": "https://arxiv.org/abs/2205.01068v4" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Punit Singh Koura", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "formulator" ], "source": { "citation": "Zhang and colleagues: OPT open pretrained transformer family — original paper", "url": "https://arxiv.org/abs/2205.01068v4" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Anjali Sridhar", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "formulator" ], "source": { "citation": "Zhang and colleagues: OPT open pretrained transformer family — original paper", "url": "https://arxiv.org/abs/2205.01068v4" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Tianlu Wang", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "formulator" ], "source": { "citation": "Zhang and colleagues: OPT open pretrained transformer family — original paper", "url": "https://arxiv.org/abs/2205.01068v4" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Luke Zettlemoyer", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "formulator" ], "source": { "citation": "Zhang and colleagues: OPT open pretrained transformer family — original paper", "url": "https://arxiv.org/abs/2205.01068v4" } } ], "evidence": [ { "description": "Bibliographic/stated-result source; not a WRS proof verification", "id": "source-0", "kind": "source_reference", "source": { "citation": "Zhang and colleagues: OPT open pretrained transformer family — original paper", "url": "https://arxiv.org/abs/2205.01068v4" } }, { "description": "Bibliographic/stated-result source; not a WRS proof verification", "id": "source-1", "kind": "source_reference", "source": { "citation": "Zhang and colleagues: OPT open pretrained transformer family — official company source", "url": "https://ai.meta.com/blog/democratizing-access-to-large-scale-language-models-with-opt-175b/" } } ], "origin": "HUMAN_LEGACY", "record_profile": "wrs-local-research-notes-0.1", "record_type": "research_note", "rrp_version": "0.1", "scope": { "description": "This is the 2022 OPT paper and release, not an invention of decoder-only transformers or a claim of unrestricted access to every model.", "domain": "Foundation language models" }, "sources": [ { "citation": "Zhang and colleagues: OPT open pretrained transformer family — original paper", "url": "https://arxiv.org/abs/2205.01068v4" }, { "citation": "Zhang and colleagues: OPT open pretrained transformer family — official company source", "url": "https://ai.meta.com/blog/democratizing-access-to-large-scale-language-models-with-opt-175b/" } ], "statement": "# Zhang and colleagues: OPT open pretrained transformer family\n\nReleased a family of 125M–175B-parameter decoder-only language models with training logs and evaluation, expanding access to large-scale model research.\n\n## Scientific scope\n\nThis is the 2022 OPT paper and release, not an invention of decoder-only transformers or a claim of unrestricted access to every model.\n\n## Reading status\n\nThis is a source-linked WRS catalogue note. A full explanatory guide and independent scientific reconstruction have not been completed for this entry. Read the identified original work for its complete narrative and arguments.\n\n## People and systems\n\nSusan Zhang; Stephen Roller; Naman Goyal; Mikel Artetxe; Moya Chen; Shuohui Chen; Christopher Dewan; Mona Diab; Xian Li; Xi Victoria Lin; Todor Mihaylov; Myle Ott; Sam Shleifer; Kurt Shuster; Daniel Simig; Punit Singh Koura; Anjali Sridhar; Tianlu Wang; Luke Zettlemoyer\n\nComplete named paper byline. Human research, source-reported system execution and experimental validation are distinct; individual effort shares are not measured.\n\n\n\n## Original sources\n\n- [Zhang and colleagues: OPT open pretrained transformer family — original paper](https://arxiv.org/abs/2205.01068v4)\n- [Zhang and colleagues: OPT open pretrained transformer family — official company source](https://ai.meta.com/blog/democratizing-access-to-large-scale-language-models-with-opt-175b/)\n", "summary": "Released a family of 125M–175B-parameter decoder-only language models with training logs and evaluation, expanding access to large-scale model research.", "title": "Zhang and colleagues: OPT open pretrained transformer family" }, "record_id": "urn:uuid:402ede42-2687-46ed-a317-75491cd789b4", "version": 1, "source_ref": { "record_id": "urn:uuid:402ede42-2687-46ed-a317-75491cd789b4", "version": 1, "content_hash": "sha256:7260ec5a4a2eebaa9f0d2b996a5f74eb1ea16426ee7c485508a8a05a6d66084c" }, "export_view": "wrs-public-editorial-view-1", "public_view_hash": "sha256:f4a88d604f1589301719793bb26c84f602d943f158d4bfff5d14cf4ab2601867" }, "target": { "record_id": "urn:uuid:402ede42-2687-46ed-a317-75491cd789b4", "version": 1, "content_hash": "sha256:7260ec5a4a2eebaa9f0d2b996a5f74eb1ea16426ee7c485508a8a05a6d66084c" }, "published_grade": { "profile": "wrs-published-grade-1", "edition": "2026-10-28.1", "id": "zhang-et-al-opt-2022", "entry_id": "zhang-et-al-opt-2022", "revision": 1, "title": "Zhang and colleagues: OPT open pretrained transformer family", "context": "Foundation language models", "track": "Human-designed AI research method or resource", "band": "B", "modifier": null, "status": "PROVISIONAL_EDITORIAL", "assessed_on": "2026-09-27", "evidence_cutoff": "2026-09-27", "achievement_state": "SUPPORTED_SCOPED_RESULT", "what_changed": "Released a family of 125M–175B-parameter decoder-only language models with training logs and evaluation, expanding access to large-scale model research.", "scope": "Released a family of 125M–175B-parameter decoder-only language models with training logs and evaluation, expanding access to large-scale model research. Scope limitation: This is the 2022 OPT paper and release, not an invention of decoder-only transformers or a claim of unrestricted access to every model.", "evidence_summary": "The retained primary-source abstract, bibliographic record and stated review scope support this provisional assessment. No independent experimental replication by WRS is claimed. This is the 2022 OPT paper and release, not an invention of decoder-only transformers or a claim of unrestricted access to every model.", "references": [ { "title": "Zhang and colleagues: OPT open pretrained transformer family — original paper", "url": "https://arxiv.org/abs/2205.01068v4", "locator": "Abstract and complete bibliographic byline for 2205.01068v4; not an independent proof or experimental replication." }, { "title": "Zhang and colleagues: OPT open pretrained transformer family — official company source", "url": "https://ai.meta.com/blog/democratizing-access-to-large-scale-language-models-with-opt-175b/", "locator": "Source-linked contribution and stated scientific limits; not an independent replication." } ], "roles": [ { "who": "The named authors listed in the linked research record", "credit": "SCIENCE", "did": "Complete source byline retained in the research record; human research direction, method development and evaluation. Individual labor shares are not measured." } ], "target": { "record_id": "urn:uuid:402ede42-2687-46ed-a317-75491cd789b4", "version": 1, "content_hash": "sha256:7260ec5a4a2eebaa9f0d2b996a5f74eb1ea16426ee7c485508a8a05a6d66084c" }, "scientific_correctness_certified": false, "independent_expert_calibration": false, "method": "PROPRIETARY", "points": "18000" }, "canonical_agent_submission": false, "scientific_verification_awarded": false, "reader_guide": null, "reading_status": "SOURCE_LINKED_NOTE" }