{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/stealing-part-of-a-production-language-model","title":"Stealing Part of a Production Language Model","arxiv_id":"2403.06634","date":"2024-03-11","proceeding":null,"authors":["Nicholas Carlini","Daniel Paleka","Krishnamurthy Dj Dvijotham","Thomas Steinke","Jonathan Hayase","A. Feder Cooper","Katherine Lee","Matthew Jagielski","Milad Nasr","Arthur Conmy","Itay Yona","Eric Wallace","David Rolnick","Florian Tramèr"],"abstract":"We introduce the first model-stealing attack that extracts precise, nontrivial information from black-box production language models like OpenAI's ChatGPT or Google's PaLM-2. Specifically, our attack recovers the embedding projection layer (up to symmetries) of a transformer model, given typical API access. For under \\$20 USD, our attack extracts the entire projection matrix of OpenAI's Ada and Babbage language models. We thereby confirm, for the first time, that these black-box models have a hidden dimension of 1024 and 2048, respectively. We also recover the exact hidden dimension size of the gpt-3.5-turbo model, and estimate it would cost under $2,000 in queries to recover the entire projection matrix. We conclude with potential defenses and mitigations, and discuss the implications of possible future work that could extend our attack.","url_abs":"https://arxiv.org/abs/2403.06634v2","url_pdf":"https://arxiv.org/pdf/2403.06634v2.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"links_only","authors_date_abstract":"arXiv metadata, CC0 1.0 (https://info.arxiv.org/help/license), from the Kaggle arXiv metadata snapshot of 2026-09-12"},"code_links":[{"paper_slug":"stealing-part-of-a-production-language-model","repo_url":"https://github.com/dpaleka/stealing-part-lm-supplementary","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":null}],"tasks":[],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/2403.06634","atlas_url":"https://app.syntology.ai/?focus=2403.06634","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2403.06634"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-25T09:33:49+00:00","read_at_is":"when the build read Syntology's graph, not when any sample ran","claim":"Per-sample execution status on synthesized fixtures; not a correctness claim about the paper. Samples come from repositories linked to the paper, official or community; repo_kind says which.","repos":[{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/dpaleka/stealing-part-lm-supplementary","reach":null}],"summary":{"ran_honours":1,"ran_fixture":2,"ran_draft_wrong":1},"by_repo_kind":{"listed":{"samples":4,"ran":4,"repositories":1}},"repo_kind_vocabulary":{"official":"The archive marks this repository official for the paper","named_in_paper":"The archive records that the paper mentions this repository; it is not marked official","listed":"In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper","found_in_text":"Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted","community":"Not in the archive's code links for this paper; a community repository Syntology harvested"},"n_pointer_only_for_licence":0,"samples":[{"code_sha256_prefix":"94df5abf385c6b88","entry":"carlini_error","repo":"dpaleka/stealing-part-lm-supplementary","repo_kind":"listed","path":"optimize_logit_queries/run_attack.py","file_url":"https://github.com/dpaleka/stealing-part-lm-supplementary/blob/HEAD/optimize_logit_queries/run_attack.py","link_basis":"first_harvest_node","language":"python","status":"ran_honours","verification_level":1,"contract_check":"HONOURS","metamorphic_tier":"invariant","behaviour_fingerprint":true,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"94df5abf385c6b88"}},{"code_sha256_prefix":"1409088ddeb0b23c","entry":"is_proper","repo":"dpaleka/stealing-part-lm-supplementary","repo_kind":"listed","path":"optimize_logit_queries/run_attack.py","file_url":"https://github.com/dpaleka/stealing-part-lm-supplementary/blob/HEAD/optimize_logit_queries/run_attack.py","link_basis":"first_harvest_node","language":"python","status":"ran_fixture","verification_level":1,"contract_check":"RAISES","metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"1409088ddeb0b23c"}},{"code_sha256_prefix":"5225f4c99476dfaf","entry":"make_chat_logprob_dict_from_completion_data","repo":"dpaleka/stealing-part-lm-supplementary","repo_kind":"listed","path":"query_logprobs_emulator/query_logprobs.py","file_url":"https://github.com/dpaleka/stealing-part-lm-supplementary/blob/HEAD/query_logprobs_emulator/query_logprobs.py","link_basis":"first_harvest_node","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"5225f4c99476dfaf"}},{"code_sha256_prefix":"3665c1c5ad3a4358","entry":"make_proper","repo":"dpaleka/stealing-part-lm-supplementary","repo_kind":"listed","path":"optimize_logit_queries/run_attack.py","file_url":"https://github.com/dpaleka/stealing-part-lm-supplementary/blob/HEAD/optimize_logit_queries/run_attack.py","link_basis":"first_harvest_node","language":"python","status":"ran_fixture","verification_level":1,"contract_check":"RAISES","metamorphic_tier":"invariant","behaviour_fingerprint":true,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"3665c1c5ad3a4358"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}