{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:IGKT6SRZTGU5DDTJJRTERXAYX4","short_pith_number":"pith:IGKT6SRZ","schema_version":"1.0","canonical_sha256":"41953f4a3999a9d18e694c6648dc18bf2304b9d998d40c01bca90fd75ae303db","source":{"kind":"arxiv","id":"2401.16818","version":2},"attestation_state":"computed","paper":{"title":"H2O-Danube-1.8B Technical Report","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Gabor Fodor, Maximilian Jeblick, Nischay Dhankhar, Pascal Pfeiffer, Philipp Singer, Sri Satish Ambati, Yauhen Babakhin","submitted_at":"2024-01-30T08:45:08Z","abstract_excerpt":"We present H2O-Danube, a series of small 1.8B language models consisting of H2O-Danube-1.8B, trained on 1T tokens, and the incremental improved H2O-Danube2-1.8B trained on an additional 2T tokens. Our models exhibit highly competitive metrics across a multitude of benchmarks and, as of the time of this writing, H2O-Danube2-1.8B achieves the top ranking on Open LLM Leaderboard for all models below the 2B parameter range. The models follow core principles of LLama 2 and Mistral, and we leverage and refine various techniques for pre-training large language models. We additionally release chat mod"},"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":"2401.16818","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-01-30T08:45:08Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"99b9b6d3d696a52da3dc3f24ec1a35cb3106f738ad7a1e74fcfd3b90b9c96a41","abstract_canon_sha256":"05468d882be0bb355852a825b2ef8181b0ac2561887cf59583f22a74fc42c2d4"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:08:01.531139Z","signature_b64":"eWsmZmeQBI6aODVSK/EYFKTRqjg60iiQ4LpComSapUtxe20ecSpgZL9oqSj/qnpT6YbzwkEMb4Xx0RH06GGdBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"41953f4a3999a9d18e694c6648dc18bf2304b9d998d40c01bca90fd75ae303db","last_reissued_at":"2026-07-05T08:08:01.530671Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:08:01.530671Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"H2O-Danube-1.8B Technical Report","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Gabor Fodor, Maximilian Jeblick, Nischay Dhankhar, Pascal Pfeiffer, Philipp Singer, Sri Satish Ambati, Yauhen Babakhin","submitted_at":"2024-01-30T08:45:08Z","abstract_excerpt":"We present H2O-Danube, a series of small 1.8B language models consisting of H2O-Danube-1.8B, trained on 1T tokens, and the incremental improved H2O-Danube2-1.8B trained on an additional 2T tokens. Our models exhibit highly competitive metrics across a multitude of benchmarks and, as of the time of this writing, H2O-Danube2-1.8B achieves the top ranking on Open LLM Leaderboard for all models below the 2B parameter range. The models follow core principles of LLama 2 and Mistral, and we leverage and refine various techniques for pre-training large language models. We additionally release chat mod"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2401.16818","kind":"arxiv","version":2},"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/2401.16818/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":"2401.16818","created_at":"2026-07-05T08:08:01.530728+00:00"},{"alias_kind":"arxiv_version","alias_value":"2401.16818v2","created_at":"2026-07-05T08:08:01.530728+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2401.16818","created_at":"2026-07-05T08:08:01.530728+00:00"},{"alias_kind":"pith_short_12","alias_value":"IGKT6SRZTGU5","created_at":"2026-07-05T08:08:01.530728+00:00"},{"alias_kind":"pith_short_16","alias_value":"IGKT6SRZTGU5DDTJ","created_at":"2026-07-05T08:08:01.530728+00:00"},{"alias_kind":"pith_short_8","alias_value":"IGKT6SRZ","created_at":"2026-07-05T08:08:01.530728+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2502.02737","citing_title":"SmolLM2: When Smol Goes Big -- Data-Centric Training of a Small Language Model","ref_index":223,"is_internal_anchor":false},{"citing_arxiv_id":"2605.07783","citing_title":"Chain-based Distillation for Effective Initialization of Variable-Sized Small Language Models","ref_index":96,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/IGKT6SRZTGU5DDTJJRTERXAYX4","json":"https://pith.science/pith/IGKT6SRZTGU5DDTJJRTERXAYX4.json","graph_json":"https://pith.science/api/pith-number/IGKT6SRZTGU5DDTJJRTERXAYX4/graph.json","events_json":"https://pith.science/api/pith-number/IGKT6SRZTGU5DDTJJRTERXAYX4/events.json","paper":"https://pith.science/paper/IGKT6SRZ"},"agent_actions":{"view_html":"https://pith.science/pith/IGKT6SRZTGU5DDTJJRTERXAYX4","download_json":"https://pith.science/pith/IGKT6SRZTGU5DDTJJRTERXAYX4.json","view_paper":"https://pith.science/paper/IGKT6SRZ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2401.16818&json=true","fetch_graph":"https://pith.science/api/pith-number/IGKT6SRZTGU5DDTJJRTERXAYX4/graph.json","fetch_events":"https://pith.science/api/pith-number/IGKT6SRZTGU5DDTJJRTERXAYX4/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/IGKT6SRZTGU5DDTJJRTERXAYX4/action/timestamp_anchor","attest_storage":"https://pith.science/pith/IGKT6SRZTGU5DDTJJRTERXAYX4/action/storage_attestation","attest_author":"https://pith.science/pith/IGKT6SRZTGU5DDTJJRTERXAYX4/action/author_attestation","sign_citation":"https://pith.science/pith/IGKT6SRZTGU5DDTJJRTERXAYX4/action/citation_signature","submit_replication":"https://pith.science/pith/IGKT6SRZTGU5DDTJJRTERXAYX4/action/replication_record"}},"created_at":"2026-07-05T08:08:01.530728+00:00","updated_at":"2026-07-05T08:08:01.530728+00:00"}