{"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/observational-scaling-laws-and-the","title":"Observational Scaling Laws and the Predictability of Language Model Performance","arxiv_id":"2405.10938","date":"2024-05-17","proceeding":null,"authors":["Yangjun Ruan","Chris J. Maddison","Tatsunori Hashimoto"],"abstract":"Understanding how language model performance varies with scale is critical to benchmark and algorithm development. Scaling laws are one approach to building this understanding, but the requirement of training models across many different scales has limited their use. We propose an alternative, observational approach that bypasses model training and instead builds scaling laws from ~100 publically available models. Building a single scaling law from multiple model families is challenging due to large variations in their training compute efficiencies and capabilities. However, we show that these variations are consistent with a simple, generalized scaling law where language model performance is a function of a low-dimensional capability space, and model families only vary in their efficiency in converting training compute to capabilities. Using this approach, we show the surprising predictability of complex scaling phenomena: we show that several emergent phenomena follow a smooth, sigmoidal behavior and are predictable from small models; we show that the agent performance of models such as GPT-4 can be precisely predicted from simpler non-agentic benchmarks; and we show how to predict the impact of post-training interventions like Chain-of-Thought and Self-Consistency as language model capabilities continue to improve.","url_abs":"https://arxiv.org/abs/2405.10938v3","url_pdf":"https://arxiv.org/pdf/2405.10938v3.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":"observational-scaling-laws-and-the","repo_url":"https://github.com/ryoungj/obsscaling","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok","spdx":"Apache-2.0"}}],"tasks":[{"task_slug":"language-modeling","task_name":"Language Modeling"},{"task_slug":"language-modelling","task_name":"Language Modelling"}],"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":{"atlas_url":"https://app.syntology.ai/?focus=2405.10938","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2405.10938"}},"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":"deterministic:regex_extraction","url":"https://github.com/ryoungj/ObsScaling","reach":{"status":"ok","spdx":"Apache-2.0"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/ryoungj/obsscaling","reach":{"status":"ok","spdx":"Apache-2.0"}}],"summary":{"ran":7,"unverified":3},"by_repo_kind":{"official":{"samples":10,"ran":7,"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":"cb1b15824bb4b525","entry":"compute_object_val","repo":"ryoungj/ObsScaling","repo_kind":"official","path":"utils/selection.py","file_url":"https://github.com/ryoungj/ObsScaling/blob/HEAD/utils/selection.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"cb1b15824bb4b525"}},{"code_sha256_prefix":"6cd43bfadf05bc38","entry":"format_x_metric_names","repo":"ryoungj/ObsScaling","repo_kind":"official","path":"utils/plot.py","file_url":"https://github.com/ryoungj/ObsScaling/blob/HEAD/utils/plot.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"6cd43bfadf05bc38"}},{"code_sha256_prefix":"be5e108bf6b36aa5","entry":"get_model_family","repo":"ryoungj/ObsScaling","repo_kind":"official","path":"utils/selection.py","file_url":"https://github.com/ryoungj/ObsScaling/blob/HEAD/utils/selection.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"be5e108bf6b36aa5"}},{"code_sha256_prefix":"ebbcecaf80a2fb9b","entry":"get_selected_df","repo":"ryoungj/ObsScaling","repo_kind":"official","path":"utils/selection.py","file_url":"https://github.com/ryoungj/ObsScaling/blob/HEAD/utils/selection.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"ebbcecaf80a2fb9b"}},{"code_sha256_prefix":"686846dcc20a9e68","entry":"pca_impute","repo":"ryoungj/ObsScaling","repo_kind":"official","path":"utils/data.py","file_url":"https://github.com/ryoungj/ObsScaling/blob/HEAD/utils/data.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"686846dcc20a9e68"}},{"code_sha256_prefix":"bfcb571bde2ee367","entry":"perform_pca","repo":"ryoungj/ObsScaling","repo_kind":"official","path":"utils/data.py","file_url":"https://github.com/ryoungj/ObsScaling/blob/HEAD/utils/data.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"bfcb571bde2ee367"}},{"code_sha256_prefix":"fb1508814f95a92b","entry":"remove_code_models","repo":"ryoungj/ObsScaling","repo_kind":"official","path":"utils/constants.py","file_url":"https://github.com/ryoungj/ObsScaling/blob/HEAD/utils/constants.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"fb1508814f95a92b"}},{"code_sha256_prefix":"719b3607af459254","entry":"keep_code_models","repo":"ryoungj/ObsScaling","repo_kind":"official","path":"utils/constants.py","file_url":"https://github.com/ryoungj/ObsScaling/blob/HEAD/utils/constants.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"719b3607af459254"}},{"code_sha256_prefix":"7beb1b630fd725dd","entry":"load_base_llm_benchmark_eval","repo":"ryoungj/ObsScaling","repo_kind":"official","path":"utils/helper.py","file_url":"https://github.com/ryoungj/ObsScaling/blob/HEAD/utils/helper.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"7beb1b630fd725dd"}},{"code_sha256_prefix":"c9186d442812ca7f","entry":"pca_preprocess_data","repo":"ryoungj/ObsScaling","repo_kind":"official","path":"utils/data.py","file_url":"https://github.com/ryoungj/ObsScaling/blob/HEAD/utils/data.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"c9186d442812ca7f"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}