{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:E7DBHNFUDXS34D3L4MZBLQICX5","short_pith_number":"pith:E7DBHNFU","schema_version":"1.0","canonical_sha256":"27c613b4b41de5be0f6be33215c102bf76cc9425e4fc74e63a015584069ebe42","source":{"kind":"arxiv","id":"2309.11568","version":1},"attestation_state":"computed","paper":{"title":"BTLM-3B-8K: 7B Parameter Performance in a 3B Parameter Model","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL","cs.LG"],"primary_cat":"cs.AI","authors_text":"Bowen Yang, Daria Soboleva, Faisal Al-Khateeb, Hemant Khachane, Jacob Robert Steeves, Joel Hestness, Marvin Tom, Natalia Vassilieva, Nolan Dey, Ribhu Pathria, Robert Myers, Shaheer Muhammad, Zhiming (Charles) Chen","submitted_at":"2023-09-20T18:12:56Z","abstract_excerpt":"We introduce the Bittensor Language Model, called \"BTLM-3B-8K\", a new state-of-the-art 3 billion parameter open-source language model. BTLM-3B-8K was trained on 627B tokens from the SlimPajama dataset with a mixture of 2,048 and 8,192 context lengths. BTLM-3B-8K outperforms all existing 3B parameter models by 2-5.5% across downstream tasks. BTLM-3B-8K is even competitive with some 7B parameter models. Additionally, BTLM-3B-8K provides excellent long context performance, outperforming MPT-7B-8K and XGen-7B-8K on tasks up to 8,192 context length. We trained the model on a cleaned and deduplicate"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2309.11568","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2023-09-20T18:12:56Z","cross_cats_sorted":["cs.CL","cs.LG"],"title_canon_sha256":"c951520fc3ab1387f31b459e66a82f1e7e82611cb20745368281ba8add37ca89","abstract_canon_sha256":"83390a7ad4b352946b968996e3aa38c0eb421ecbcc5301673bba2e41d8423758"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:52:51.717618Z","signature_b64":"ZwmBg+8Td9lu8U+jTxNXwtHAzKXbEmHEDJKSoqmTjonVC+Gho+D0A4og88KaHoKmaQiDz4DPbi3bf4ZR9K2TBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"27c613b4b41de5be0f6be33215c102bf76cc9425e4fc74e63a015584069ebe42","last_reissued_at":"2026-07-05T06:52:51.717147Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:52:51.717147Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"BTLM-3B-8K: 7B Parameter Performance in a 3B Parameter Model","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL","cs.LG"],"primary_cat":"cs.AI","authors_text":"Bowen Yang, Daria Soboleva, Faisal Al-Khateeb, Hemant Khachane, Jacob Robert Steeves, Joel Hestness, Marvin Tom, Natalia Vassilieva, Nolan Dey, Ribhu Pathria, Robert Myers, Shaheer Muhammad, Zhiming (Charles) Chen","submitted_at":"2023-09-20T18:12:56Z","abstract_excerpt":"We introduce the Bittensor Language Model, called \"BTLM-3B-8K\", a new state-of-the-art 3 billion parameter open-source language model. BTLM-3B-8K was trained on 627B tokens from the SlimPajama dataset with a mixture of 2,048 and 8,192 context lengths. BTLM-3B-8K outperforms all existing 3B parameter models by 2-5.5% across downstream tasks. BTLM-3B-8K is even competitive with some 7B parameter models. Additionally, BTLM-3B-8K provides excellent long context performance, outperforming MPT-7B-8K and XGen-7B-8K on tasks up to 8,192 context length. We trained the model on a cleaned and deduplicate"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2309.11568","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/2309.11568/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"},"aliases":[{"alias_kind":"arxiv","alias_value":"2309.11568","created_at":"2026-07-05T06:52:51.717206+00:00"},{"alias_kind":"arxiv_version","alias_value":"2309.11568v1","created_at":"2026-07-05T06:52:51.717206+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2309.11568","created_at":"2026-07-05T06:52:51.717206+00:00"},{"alias_kind":"pith_short_12","alias_value":"E7DBHNFUDXS3","created_at":"2026-07-05T06:52:51.717206+00:00"},{"alias_kind":"pith_short_16","alias_value":"E7DBHNFUDXS34D3L","created_at":"2026-07-05T06:52:51.717206+00:00"},{"alias_kind":"pith_short_8","alias_value":"E7DBHNFU","created_at":"2026-07-05T06:52:51.717206+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.06330","citing_title":"Fine-Tuning Small Language Models for Solution-Oriented Windows Event Log Analysis","ref_index":6,"is_internal_anchor":false},{"citing_arxiv_id":"2604.04738","citing_title":"Fine-Tuning Integrity for Modern Neural Networks: Structured Drift Proofs via Norm, Rank, and Sparsity Certificates","ref_index":10,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/E7DBHNFUDXS34D3L4MZBLQICX5","json":"https://pith.science/pith/E7DBHNFUDXS34D3L4MZBLQICX5.json","graph_json":"https://pith.science/api/pith-number/E7DBHNFUDXS34D3L4MZBLQICX5/graph.json","events_json":"https://pith.science/api/pith-number/E7DBHNFUDXS34D3L4MZBLQICX5/events.json","paper":"https://pith.science/paper/E7DBHNFU"},"agent_actions":{"view_html":"https://pith.science/pith/E7DBHNFUDXS34D3L4MZBLQICX5","download_json":"https://pith.science/pith/E7DBHNFUDXS34D3L4MZBLQICX5.json","view_paper":"https://pith.science/paper/E7DBHNFU","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2309.11568&json=true","fetch_graph":"https://pith.science/api/pith-number/E7DBHNFUDXS34D3L4MZBLQICX5/graph.json","fetch_events":"https://pith.science/api/pith-number/E7DBHNFUDXS34D3L4MZBLQICX5/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/E7DBHNFUDXS34D3L4MZBLQICX5/action/timestamp_anchor","attest_storage":"https://pith.science/pith/E7DBHNFUDXS34D3L4MZBLQICX5/action/storage_attestation","attest_author":"https://pith.science/pith/E7DBHNFUDXS34D3L4MZBLQICX5/action/author_attestation","sign_citation":"https://pith.science/pith/E7DBHNFUDXS34D3L4MZBLQICX5/action/citation_signature","submit_replication":"https://pith.science/pith/E7DBHNFUDXS34D3L4MZBLQICX5/action/replication_record"}},"created_at":"2026-07-05T06:52:51.717206+00:00","updated_at":"2026-07-05T06:52:51.717206+00:00"}