{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:JBSZIZ2AE3VZ6KOLJJVP5KM4CU","short_pith_number":"pith:JBSZIZ2A","schema_version":"1.0","canonical_sha256":"486594674026eb9f29cb4a6afea99c1538f2924fedce1a64515528bf93ea8fd0","source":{"kind":"arxiv","id":"2606.22189","version":1},"attestation_state":"computed","paper":{"title":"L20-Edu-135M: An Auditable Single-GPU Study of Data-Efficient Small Language Modeling","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Yin Li","submitted_at":"2026-06-20T18:42:37Z","abstract_excerpt":"Small language models are cheap to serve and feasible on local hardware, but strong public 135M-class systems are commonly trained with hundreds of billions to trillions of tokens on large clusters. We study a sharply resource-constrained regime: a complete 134.5M-parameter language-model pipeline executed on one NVIDIA L20 GPU. The released checkpoint, L20-Edu-135M, receives approximately 13B pretraining tokens: 10B FineWeb-Edu tokens followed by a 3B-token educational, mathematics, code, and reasoning mixture. We document the architecture, data gates, cross-source MinHash/LSH near-deduplicat"},"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":"2606.22189","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-06-20T18:42:37Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"31958fe0ec59ae65c9b8d8d7a673d5a0a2b6a9b96b42ff88dd94a596c75861b4","abstract_canon_sha256":"d00710a258b66fdef6a644ff639a5dc2a4b1f49f075d4a612379ea85c8413aeb"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-23T02:13:31.447522Z","signature_b64":"BhO14TYXKmRe6V+cKH/TaHPxlwjMwgXPNu/uCINE5UqIBTqVZ7gM3xAtE+sLCvN+Dl5ae3BDPJkZLBoNm4osBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"486594674026eb9f29cb4a6afea99c1538f2924fedce1a64515528bf93ea8fd0","last_reissued_at":"2026-06-23T02:13:31.447046Z","signature_status":"signed_v1","first_computed_at":"2026-06-23T02:13:31.447046Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"L20-Edu-135M: An Auditable Single-GPU Study of Data-Efficient Small Language Modeling","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Yin Li","submitted_at":"2026-06-20T18:42:37Z","abstract_excerpt":"Small language models are cheap to serve and feasible on local hardware, but strong public 135M-class systems are commonly trained with hundreds of billions to trillions of tokens on large clusters. We study a sharply resource-constrained regime: a complete 134.5M-parameter language-model pipeline executed on one NVIDIA L20 GPU. The released checkpoint, L20-Edu-135M, receives approximately 13B pretraining tokens: 10B FineWeb-Edu tokens followed by a 3B-token educational, mathematics, code, and reasoning mixture. We document the architecture, data gates, cross-source MinHash/LSH near-deduplicat"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.22189","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/2606.22189/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":"2606.22189","created_at":"2026-06-23T02:13:31.447104+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.22189v1","created_at":"2026-06-23T02:13:31.447104+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.22189","created_at":"2026-06-23T02:13:31.447104+00:00"},{"alias_kind":"pith_short_12","alias_value":"JBSZIZ2AE3VZ","created_at":"2026-06-23T02:13:31.447104+00:00"},{"alias_kind":"pith_short_16","alias_value":"JBSZIZ2AE3VZ6KOL","created_at":"2026-06-23T02:13:31.447104+00:00"},{"alias_kind":"pith_short_8","alias_value":"JBSZIZ2A","created_at":"2026-06-23T02:13:31.447104+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/JBSZIZ2AE3VZ6KOLJJVP5KM4CU","json":"https://pith.science/pith/JBSZIZ2AE3VZ6KOLJJVP5KM4CU.json","graph_json":"https://pith.science/api/pith-number/JBSZIZ2AE3VZ6KOLJJVP5KM4CU/graph.json","events_json":"https://pith.science/api/pith-number/JBSZIZ2AE3VZ6KOLJJVP5KM4CU/events.json","paper":"https://pith.science/paper/JBSZIZ2A"},"agent_actions":{"view_html":"https://pith.science/pith/JBSZIZ2AE3VZ6KOLJJVP5KM4CU","download_json":"https://pith.science/pith/JBSZIZ2AE3VZ6KOLJJVP5KM4CU.json","view_paper":"https://pith.science/paper/JBSZIZ2A","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.22189&json=true","fetch_graph":"https://pith.science/api/pith-number/JBSZIZ2AE3VZ6KOLJJVP5KM4CU/graph.json","fetch_events":"https://pith.science/api/pith-number/JBSZIZ2AE3VZ6KOLJJVP5KM4CU/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/JBSZIZ2AE3VZ6KOLJJVP5KM4CU/action/timestamp_anchor","attest_storage":"https://pith.science/pith/JBSZIZ2AE3VZ6KOLJJVP5KM4CU/action/storage_attestation","attest_author":"https://pith.science/pith/JBSZIZ2AE3VZ6KOLJJVP5KM4CU/action/author_attestation","sign_citation":"https://pith.science/pith/JBSZIZ2AE3VZ6KOLJJVP5KM4CU/action/citation_signature","submit_replication":"https://pith.science/pith/JBSZIZ2AE3VZ6KOLJJVP5KM4CU/action/replication_record"}},"created_at":"2026-06-23T02:13:31.447104+00:00","updated_at":"2026-06-23T02:13:31.447104+00:00"}