{"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/understanding-the-effectiveness-of-early","title":"Early Weight Averaging meets High Learning Rates for LLM Pre-training","arxiv_id":"2306.03241","date":"2023-06-05","proceeding":null,"authors":["Sunny Sanyal","Atula Neerkaje","Jean Kaddour","Abhishek Kumar","Sujay Sanghavi"],"abstract":"Training Large Language Models (LLMs) incurs significant cost; hence, any strategy that accelerates model convergence is helpful. In this paper, we investigate the ability of a simple idea checkpoint averaging along the trajectory of a training run to improve both convergence and generalization quite early on during training. Here we show that models trained with high learning rates observe higher gains due to checkpoint averaging. Furthermore, these gains are amplified when checkpoints are sampled with considerable spacing in training steps. Our training recipe outperforms conventional training and popular checkpoint averaging baselines such as exponential moving average (EMA) and stochastic moving average (SWA). We evaluate our training recipe by pre-training LLMs, where high learning rates are inherently preferred due to extremely large batch sizes. Specifically, we pre-trained nanoGPT-2 models of varying sizes, small (125M), medium (335M), and large (770M)on the OpenWebText dataset, comprised of 9B tokens. Additionally, we present results for publicly available Pythia LLMs, ranging from 1B to 12B, which were trained on the PILE-deduped dataset containing 207B tokens.","url_abs":"https://arxiv.org/abs/2306.03241v2","url_pdf":"https://arxiv.org/pdf/2306.03241v2.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":"understanding-the-effectiveness-of-early","repo_url":"https://github.com/sanyalsunny111/early_weight_avg","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}}],"tasks":[],"methods":[{"method_slug":"pythia","method_name":"Pythia"},{"method_slug":"test","method_name":"Test"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/2306.03241","atlas_url":"https://app.syntology.ai/?focus=2306.03241","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2306.03241"}},"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":"deterministic:regex_extraction","url":"https://github.com/sanyalsunny111/Early_Weight_Avg","reach":{"status":"ok"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/sanyalsunny111/early_weight_avg","reach":{"status":"ok"}}],"summary":{"ran_honours":1,"ran_violates":1,"ran_draft_wrong":1},"by_repo_kind":{"official":{"samples":3,"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":3,"samples":[{"code_sha256_prefix":"75f5853321b37296","entry":"extract_number","repo":"sanyalsunny111/early_weight_avg","repo_kind":"official","path":"nanoGPT2-Experiments/lawa.py","file_url":"https://github.com/sanyalsunny111/early_weight_avg/blob/HEAD/nanoGPT2-Experiments/lawa.py","link_basis":"first_harvest_node","language":"python","status":"ran_honours","verification_level":2,"contract_check":"HONOURS","metamorphic_tier":"well_formed","behaviour_fingerprint":true,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"75f5853321b37296"}},{"code_sha256_prefix":"1ed19b773b317db5","entry":"is_float_dtype","repo":"sanyalsunny111/early_weight_avg","repo_kind":"official","path":"nanoGPT2-Experiments/lawa.py","file_url":"https://github.com/sanyalsunny111/early_weight_avg/blob/HEAD/nanoGPT2-Experiments/lawa.py","link_basis":"first_harvest_node","language":"python","status":"ran_violates","verification_level":1,"contract_check":"VIOLATES","metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"1ed19b773b317db5"}},{"code_sha256_prefix":"d0aa4e90fc5641bb","entry":"uni_update","repo":"sanyalsunny111/early_weight_avg","repo_kind":"official","path":"nanoGPT2-Experiments/lawa.py","file_url":"https://github.com/sanyalsunny111/early_weight_avg/blob/HEAD/nanoGPT2-Experiments/lawa.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":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"d0aa4e90fc5641bb"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}