{"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/opening-the-ai-black-box-program-synthesis","title":"Opening the AI black box: program synthesis via mechanistic interpretability","arxiv_id":"2402.05110","date":"2024-02-07","proceeding":null,"authors":["Eric J. Michaud","Isaac Liao","Vedang Lad","Ziming Liu","Anish Mudide","Chloe Loughridge","Zifan Carl Guo","Tara Rezaei Kheirkhah","Mateja Vukelić","Max Tegmark"],"abstract":"We present MIPS, a novel method for program synthesis based on automated mechanistic interpretability of neural networks trained to perform the desired task, auto-distilling the learned algorithm into Python code. We test MIPS on a benchmark of 62 algorithmic tasks that can be learned by an RNN and find it highly complementary to GPT-4: MIPS solves 32 of them, including 13 that are not solved by GPT-4 (which also solves 30). MIPS uses an integer autoencoder to convert the RNN into a finite state machine, then applies Boolean or integer symbolic regression to capture the learned algorithm. As opposed to large language models, this program synthesis technique makes no use of (and is therefore not limited by) human training data such as algorithms and code from GitHub. We discuss opportunities and challenges for scaling up this approach to make machine-learned models more interpretable and trustworthy.","url_abs":"https://arxiv.org/abs/2402.05110v1","url_pdf":"https://arxiv.org/pdf/2402.05110v1.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":"abstracts"},"code_links":[{"paper_slug":"opening-the-ai-black-box-program-synthesis","repo_url":"https://github.com/ejmichaud/neural-verification","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}}],"tasks":[{"task_slug":"program-synthesis","task_name":"Program Synthesis"},{"task_slug":"symbolic-regression","task_name":"Symbolic Regression"}],"methods":[{"method_slug":"absolute-position-encodings","method_name":"Absolute Position Encodings"},{"method_slug":"adam","method_name":"Adam"},{"method_slug":"attention","method_name":"Attention"},{"method_slug":"bpe","method_name":"BPE"},{"method_slug":"dense-connections","method_name":"Dense Connections"},{"method_slug":"dropout","method_name":"Dropout"},{"method_slug":"gpt-4","method_name":"GPT-4"},{"method_slug":"label-smoothing","method_name":"Label Smoothing"},{"method_slug":"layer-normalization","method_name":"Layer Normalization"},{"method_slug":"linear-layer","method_name":"Linear Layer"},{"method_slug":"multi-head-attention","method_name":"Multi-Head Attention"},{"method_slug":"position-wise-feed-forward-layer","method_name":"Position-Wise Feed-Forward Layer"},{"method_slug":"residual-connection","method_name":"Residual Connection"},{"method_slug":"softmax","method_name":"Softmax"},{"method_slug":"transformer","method_name":"Transformer"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":null,"atlas_url":"https://app.syntology.ai/?focus=2402.05110","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2402.05110"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-24T18:15:14+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/ejmichaud/neural-verification","reach":{"status":"ok"}}],"summary":{"ran":3,"unverified":1},"by_repo_kind":{"official":{"samples":4,"ran":3,"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":4,"samples":[{"code_sha256_prefix":"b7298ebc06024744","entry":"format_prompt_for_formula","repo":"ejmichaud/neural-verification","repo_kind":"official","path":"gpt4/prompt.py","file_url":"https://github.com/ejmichaud/neural-verification/blob/HEAD/gpt4/prompt.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"b7298ebc06024744"}},{"code_sha256_prefix":"72edf25d5e85faf8","entry":"goodrepr","repo":"ejmichaud/neural-verification","repo_kind":"official","path":"normalizers/inspect_model.py","file_url":"https://github.com/ejmichaud/neural-verification/blob/HEAD/normalizers/inspect_model.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"72edf25d5e85faf8"}},{"code_sha256_prefix":"9c706360659ebe18","entry":"read_pt_file","repo":"ejmichaud/neural-verification","repo_kind":"official","path":"gpt4/prompt.py","file_url":"https://github.com/ejmichaud/neural-verification/blob/HEAD/gpt4/prompt.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"9c706360659ebe18"}},{"code_sha256_prefix":"e4e4bb276d873fd8","entry":"query_gpt","repo":"ejmichaud/neural-verification","repo_kind":"official","path":"gpt4/prompt.py","file_url":"https://github.com/ejmichaud/neural-verification/blob/HEAD/gpt4/prompt.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"e4e4bb276d873fd8"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}