{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:AUZ2KXEGQ3FFCUNGONF2K3MBBM","short_pith_number":"pith:AUZ2KXEG","canonical_record":{"source":{"id":"2412.08347","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-12-11T12:41:36Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"2a4eff7892fd81983f732ee978394e6f2775bdd4f0a5802ac013190a0dfbcb1d","abstract_canon_sha256":"ae66babe75a7f36fe26a06b25e569789c245426c6aced5ff563877cfd7a91bf2"},"schema_version":"1.0"},"canonical_sha256":"0533a55c8686ca5151a6734ba56d810b2eeb6add69cb357379b26a53780161e5","source":{"kind":"arxiv","id":"2412.08347","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2412.08347","created_at":"2026-07-05T09:47:48Z"},{"alias_kind":"arxiv_version","alias_value":"2412.08347v1","created_at":"2026-07-05T09:47:48Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2412.08347","created_at":"2026-07-05T09:47:48Z"},{"alias_kind":"pith_short_12","alias_value":"AUZ2KXEGQ3FF","created_at":"2026-07-05T09:47:48Z"},{"alias_kind":"pith_short_16","alias_value":"AUZ2KXEGQ3FFCUNG","created_at":"2026-07-05T09:47:48Z"},{"alias_kind":"pith_short_8","alias_value":"AUZ2KXEG","created_at":"2026-07-05T09:47:48Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:AUZ2KXEGQ3FFCUNGONF2K3MBBM","target":"record","payload":{"canonical_record":{"source":{"id":"2412.08347","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-12-11T12:41:36Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"2a4eff7892fd81983f732ee978394e6f2775bdd4f0a5802ac013190a0dfbcb1d","abstract_canon_sha256":"ae66babe75a7f36fe26a06b25e569789c245426c6aced5ff563877cfd7a91bf2"},"schema_version":"1.0"},"canonical_sha256":"0533a55c8686ca5151a6734ba56d810b2eeb6add69cb357379b26a53780161e5","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:47:48.508765Z","signature_b64":"qwWuN4kR7EJAwIZaQXlg7o3/uxdDqAoOelxH26J6ur/sOwAHZOLvDjfvOP2xa3UiT8vF3PjJt/DgCeaUFQEnBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"0533a55c8686ca5151a6734ba56d810b2eeb6add69cb357379b26a53780161e5","last_reissued_at":"2026-07-05T09:47:48.508370Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:47:48.508370Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2412.08347","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T09:47:48Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"RzQfAOM0berk14kasdxAVJXG0potE96+CcOz+ORuXznBfYZGIRZLwrwiLmPuqaU3Cp6rY+As1OuO005tELY3Bw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T16:57:32.819381Z"},"content_sha256":"705be3cad12f86781bb6be18b3985bb44c74eed363d818121d3e5138f67ff129","schema_version":"1.0","event_id":"sha256:705be3cad12f86781bb6be18b3985bb44c74eed363d818121d3e5138f67ff129"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:AUZ2KXEGQ3FFCUNGONF2K3MBBM","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"SmolTulu: Higher Learning Rate to Batch Size Ratios Can Lead to Better Reasoning in SLMs","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Sultan Alrashed","submitted_at":"2024-12-11T12:41:36Z","abstract_excerpt":"We present SmolTulu-1.7b-Instruct, referenced in this report as SmolTulu-DPO-1130, an instruction-tuned language model that adapts AllenAI's Tulu 3 post-training pipeline to enhance Huggingface's SmolLM2-1.7B base model. Through comprehensive empirical analysis using a 135M parameter model, we demonstrate that the relationship between learning rate and batch size significantly impacts model performance in a task-dependent manner. Our findings reveal a clear split: reasoning tasks like ARC and GSM8K benefit from higher learning rate to batch size ratios, while pattern recognition tasks such as "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2412.08347","kind":"arxiv","version":1},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2412.08347/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T09:47:48Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"DlStncMB+HzL5Tyki/pbFQqIP090bZt6hH2tclNa1UVbMHoDB+xF6D4XVFXnv35FycqMLHn0e1619JDAWvh6Aw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T16:57:32.819891Z"},"content_sha256":"7d16730710c85ee39a14198d4901e92a513660b5fee8b5bae6af14581adbdd08","schema_version":"1.0","event_id":"sha256:7d16730710c85ee39a14198d4901e92a513660b5fee8b5bae6af14581adbdd08"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/AUZ2KXEGQ3FFCUNGONF2K3MBBM/bundle.json","state_url":"https://pith.science/pith/AUZ2KXEGQ3FFCUNGONF2K3MBBM/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/AUZ2KXEGQ3FFCUNGONF2K3MBBM/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-08-08T16:57:32Z","links":{"resolver":"https://pith.science/pith/AUZ2KXEGQ3FFCUNGONF2K3MBBM","bundle":"https://pith.science/pith/AUZ2KXEGQ3FFCUNGONF2K3MBBM/bundle.json","state":"https://pith.science/pith/AUZ2KXEGQ3FFCUNGONF2K3MBBM/state.json","well_known_bundle":"https://pith.science/.well-known/pith/AUZ2KXEGQ3FFCUNGONF2K3MBBM/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:AUZ2KXEGQ3FFCUNGONF2K3MBBM","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"ae66babe75a7f36fe26a06b25e569789c245426c6aced5ff563877cfd7a91bf2","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-12-11T12:41:36Z","title_canon_sha256":"2a4eff7892fd81983f732ee978394e6f2775bdd4f0a5802ac013190a0dfbcb1d"},"schema_version":"1.0","source":{"id":"2412.08347","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2412.08347","created_at":"2026-07-05T09:47:48Z"},{"alias_kind":"arxiv_version","alias_value":"2412.08347v1","created_at":"2026-07-05T09:47:48Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2412.08347","created_at":"2026-07-05T09:47:48Z"},{"alias_kind":"pith_short_12","alias_value":"AUZ2KXEGQ3FF","created_at":"2026-07-05T09:47:48Z"},{"alias_kind":"pith_short_16","alias_value":"AUZ2KXEGQ3FFCUNG","created_at":"2026-07-05T09:47:48Z"},{"alias_kind":"pith_short_8","alias_value":"AUZ2KXEG","created_at":"2026-07-05T09:47:48Z"}],"graph_snapshots":[{"event_id":"sha256:7d16730710c85ee39a14198d4901e92a513660b5fee8b5bae6af14581adbdd08","target":"graph","created_at":"2026-07-05T09:47:48Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2412.08347/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"We present SmolTulu-1.7b-Instruct, referenced in this report as SmolTulu-DPO-1130, an instruction-tuned language model that adapts AllenAI's Tulu 3 post-training pipeline to enhance Huggingface's SmolLM2-1.7B base model. Through comprehensive empirical analysis using a 135M parameter model, we demonstrate that the relationship between learning rate and batch size significantly impacts model performance in a task-dependent manner. Our findings reveal a clear split: reasoning tasks like ARC and GSM8K benefit from higher learning rate to batch size ratios, while pattern recognition tasks such as ","authors_text":"Sultan Alrashed","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-12-11T12:41:36Z","title":"SmolTulu: Higher Learning Rate to Batch Size Ratios Can Lead to Better Reasoning in SLMs"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2412.08347","kind":"arxiv","version":1},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:705be3cad12f86781bb6be18b3985bb44c74eed363d818121d3e5138f67ff129","target":"record","created_at":"2026-07-05T09:47:48Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"ae66babe75a7f36fe26a06b25e569789c245426c6aced5ff563877cfd7a91bf2","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-12-11T12:41:36Z","title_canon_sha256":"2a4eff7892fd81983f732ee978394e6f2775bdd4f0a5802ac013190a0dfbcb1d"},"schema_version":"1.0","source":{"id":"2412.08347","kind":"arxiv","version":1}},"canonical_sha256":"0533a55c8686ca5151a6734ba56d810b2eeb6add69cb357379b26a53780161e5","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"0533a55c8686ca5151a6734ba56d810b2eeb6add69cb357379b26a53780161e5","first_computed_at":"2026-07-05T09:47:48.508370Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T09:47:48.508370Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"qwWuN4kR7EJAwIZaQXlg7o3/uxdDqAoOelxH26J6ur/sOwAHZOLvDjfvOP2xa3UiT8vF3PjJt/DgCeaUFQEnBA==","signature_status":"signed_v1","signed_at":"2026-07-05T09:47:48.508765Z","signed_message":"canonical_sha256_bytes"},"source_id":"2412.08347","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:705be3cad12f86781bb6be18b3985bb44c74eed363d818121d3e5138f67ff129","sha256:7d16730710c85ee39a14198d4901e92a513660b5fee8b5bae6af14581adbdd08"],"state_sha256":"da30120300d430a6766e7718bf1ad97159fc4a6e584aa32486c6788f8d9abae1"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Zmx2ZUJnz0fXZNUUYPOrcksBV6Y+wirqQ1CJVawahTpm7RZR8JaWLLuP7o0eIWOitTlzzUS8r4P5h48Tl5LgAQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-08T16:57:32.826081Z","bundle_sha256":"1da8799ad5d8227fe95d5f8402d57d4ed9133ea8d12c3859b1e52e2508796e5f"}}