MAG and complete-genome checkpoints are distinct
Complete genome pretraining of Bacformer
The complete-genome checkpoints are initialised from the corresponding MAG checkpoint and further pretrained on about 40,000 complete genomes from NCBI RefSeq, so the MAG and complete-genome checkpoints are different configurations of the same architecture
Recorded because results must not be moved between the MAG and complete-genome checkpoints: they differ by a further pretraining stage on a different genome type.
Evidence
Source checking verifies the cited claim or transcription. It does not establish independent reproduction.
Evidence table
Inspect claims, sources and review details
Trace each statement to its source and review. A context-only reference supports the record generally; it does not verify an individual field. Source checking does not reproduce an experiment.
One row per statement and cited source. Multiple citations are not independent evaluations. Shared locators are labelled explicitly.
21 evidence rows matching the loaded filters
| Property and statement | Original source and location | Review and provenance |
|---|---|---|
| Claimed field complete_genome_pretraining Context-only references | Bacformer repository README at commit 61dca95 Hugging Face model cards for macwiatrak/bacformer-masked-complete-genomes and macwiatrak/bacformer-large-masked-complete-genomes: 'This model was initialised from ... and further pretrained on ~40 k complete genomes' Shared locator for this statement’s cited sources; not a separate locator for each citation. Version: Commit 61dca9537429b95365d73d72a3e11c207df1ed6c | not individually reviewed No individual claim review recorded Audit detailsField: Source artifact SHA-256: Hash scope: Hash scope not separately documented; inspect source record |
| Claimed field complete_genome_pretraining Context-only references | Hugging Face model card for macwiatrak/bacformer-large-masked-complete-genomes Hugging Face model cards for macwiatrak/bacformer-masked-complete-genomes and macwiatrak/bacformer-large-masked-complete-genomes: 'This model was initialised from ... and further pretrained on ~40 k complete genomes' Shared locator for this statement’s cited sources; not a separate locator for each citation. Version: Repository revision 84f85c37a05e55559142b842c1569e265183b554 | not individually reviewed No individual claim review recorded Audit detailsField: Source artifact SHA-256: Hash scope: Hash scope not separately documented; inspect source record |
| Claimed field complete_genome_pretraining Context-only references | Hugging Face model card for macwiatrak/bacformer-masked-complete-genomes Hugging Face model cards for macwiatrak/bacformer-masked-complete-genomes and macwiatrak/bacformer-large-masked-complete-genomes: 'This model was initialised from ... and further pretrained on ~40 k complete genomes' Shared locator for this statement’s cited sources; not a separate locator for each citation. Version: Repository revision d1c461f1db75902e6502e5ba7588ea1a4ea96b29 | not individually reviewed No individual claim review recorded Audit detailsField: Source artifact SHA-256: Hash scope: Hash scope not separately documented; inspect source record |
| Rationale Recorded because results must not be moved between the MAG and complete-genome checkpoints: they differ by a further pretraining stage on a different genome type. Context-only references | Bacformer repository README at commit 61dca95 Hugging Face model cards for macwiatrak/bacformer-masked-complete-genomes and macwiatrak/bacformer-large-masked-complete-genomes: 'This model was initialised from ... and further pretrained on ~40 k complete genomes' Shared locator for this statement’s cited sources; not a separate locator for each citation. Version: Commit 61dca9537429b95365d73d72a3e11c207df1ed6c | not individually reviewed No individual claim review recorded Audit detailsField: Source artifact SHA-256: Hash scope: Hash scope not separately documented; inspect source record |
| Rationale Recorded because results must not be moved between the MAG and complete-genome checkpoints: they differ by a further pretraining stage on a different genome type. Context-only references | Hugging Face model card for macwiatrak/bacformer-large-masked-complete-genomes Hugging Face model cards for macwiatrak/bacformer-masked-complete-genomes and macwiatrak/bacformer-large-masked-complete-genomes: 'This model was initialised from ... and further pretrained on ~40 k complete genomes' Shared locator for this statement’s cited sources; not a separate locator for each citation. Version: Repository revision 84f85c37a05e55559142b842c1569e265183b554 | not individually reviewed No individual claim review recorded Audit detailsField: Source artifact SHA-256: Hash scope: Hash scope not separately documented; inspect source record |
| Rationale Recorded because results must not be moved between the MAG and complete-genome checkpoints: they differ by a further pretraining stage on a different genome type. Context-only references | Hugging Face model card for macwiatrak/bacformer-masked-complete-genomes Hugging Face model cards for macwiatrak/bacformer-masked-complete-genomes and macwiatrak/bacformer-large-masked-complete-genomes: 'This model was initialised from ... and further pretrained on ~40 k complete genomes' Shared locator for this statement’s cited sources; not a separate locator for each citation. Version: Repository revision d1c461f1db75902e6502e5ba7588ea1a4ea96b29 | not individually reviewed No individual claim review recorded Audit detailsField: Source artifact SHA-256: Hash scope: Hash scope not separately documented; inspect source record |
| Source location Hugging Face model cards for macwiatrak/bacformer-masked-complete-genomes and macwiatrak/bacformer-large-masked-complete-genomes: 'This model was initialised from ... and further pretrained on ~40 k complete genomes' Context-only references | Bacformer repository README at commit 61dca95 Hugging Face model cards for macwiatrak/bacformer-masked-complete-genomes and macwiatrak/bacformer-large-masked-complete-genomes: 'This model was initialised from ... and further pretrained on ~40 k complete genomes' Shared locator for this statement’s cited sources; not a separate locator for each citation. Version: Commit 61dca9537429b95365d73d72a3e11c207df1ed6c | not individually reviewed No individual claim review recorded Audit detailsField: Source artifact SHA-256: Hash scope: Hash scope not separately documented; inspect source record |
| Source location Hugging Face model cards for macwiatrak/bacformer-masked-complete-genomes and macwiatrak/bacformer-large-masked-complete-genomes: 'This model was initialised from ... and further pretrained on ~40 k complete genomes' Context-only references | Hugging Face model card for macwiatrak/bacformer-large-masked-complete-genomes Hugging Face model cards for macwiatrak/bacformer-masked-complete-genomes and macwiatrak/bacformer-large-masked-complete-genomes: 'This model was initialised from ... and further pretrained on ~40 k complete genomes' Shared locator for this statement’s cited sources; not a separate locator for each citation. Version: Repository revision 84f85c37a05e55559142b842c1569e265183b554 | not individually reviewed No individual claim review recorded Audit detailsField: Source artifact SHA-256: Hash scope: Hash scope not separately documented; inspect source record |
| Source location Hugging Face model cards for macwiatrak/bacformer-masked-complete-genomes and macwiatrak/bacformer-large-masked-complete-genomes: 'This model was initialised from ... and further pretrained on ~40 k complete genomes' Context-only references | Hugging Face model card for macwiatrak/bacformer-masked-complete-genomes Hugging Face model cards for macwiatrak/bacformer-masked-complete-genomes and macwiatrak/bacformer-large-masked-complete-genomes: 'This model was initialised from ... and further pretrained on ~40 k complete genomes' Shared locator for this statement’s cited sources; not a separate locator for each citation. Version: Repository revision d1c461f1db75902e6502e5ba7588ea1a4ea96b29 | not individually reviewed No individual claim review recorded Audit detailsField: Source artifact SHA-256: Hash scope: Hash scope not separately documented; inspect source record |
| Claimed value The complete-genome checkpoints are initialised from the corresponding MAG checkpoint and further pretrained on about 40,000 complete genomes from NCBI RefSeq, so the MAG and complete-genome checkpoints are different configurations of the same architecture Context-only references | Bacformer repository README at commit 61dca95 Hugging Face model cards for macwiatrak/bacformer-masked-complete-genomes and macwiatrak/bacformer-large-masked-complete-genomes: 'This model was initialised from ... and further pretrained on ~40 k complete genomes' Shared locator for this statement’s cited sources; not a separate locator for each citation. Version: Commit 61dca9537429b95365d73d72a3e11c207df1ed6c | not individually reviewed No individual claim review recorded Audit detailsField: Source artifact SHA-256: Hash scope: Hash scope not separately documented; inspect source record |
Sources and history
Release 2026-11-21-c8bea25e5abb · Record review: source checked
3 source records and release history
- Hugging Face model card for macwiatrak/bacformer-masked-complete-genomes · Original source · Repository revision d1c461f1db75902e6502e5ba7588ea1a4ea96b29
- Hugging Face model card for macwiatrak/bacformer-large-masked-complete-genomes · Original source · Repository revision 84f85c37a05e55559142b842c1569e265183b554
- Bacformer repository README at commit 61dca95 · Original source · Commit 61dca9537429b95365d73d72a3e11c207df1ed6c
Technical metadata and extraction receipts
Stable ID: bacformer-20261115-claim-bacformer-complete-genome-finetune
- areas
- microbes-communities
- contexts
- research
- field
- complete_genome_pretraining
- rationale
- Recorded because results must not be moved between the MAG and complete-genome checkpoints: they differ by a further pretraining stage on a different genome type.
- review
- artifact sha256: 02575bd9d4ae949e3c38a5a5890767b862c75bc2a20e84afa4a517bd2f067ff1; method: transcription; source-hash-verification; ai-assisted-source-review; method note: Independent review pass. Each source artifact was retrieved again at review time and its SHA-256 compared with the source record; the quoted statement was then read at the recorded locator and compared with the recorded value. Access states were rechecked rather than taken from notes.md.; note: Statement re-read against the pinned artifact by a reviewer who did not extract this batch. Every artifact except the two preprint PDFs was downloaded again at review time and hashed to exactly the value its source record records, and is byte-identical to the copy kept in artifacts/.; retrieval url: https://huggingface.co/macwiatrak/bacformer-masked-complete-genomes/raw/d1c461f1db75902e6502e5ba7588ea1a4ea96b29/README.md; reviewed at: 2026-11-15T23:55:00Z; reviewer: claude; reviewer note: Claude (Opus 5) review agent, independent of the agent that extracted this batch; no human review claimed. Transcription was checked against the pinned sources. Nothing was reproduced and no value was recomputed.
- source locator
- Hugging Face model cards for macwiatrak/bacformer-masked-complete-genomes and macwiatrak/bacformer-large-masked-complete-genomes: 'This model was initialised from ... and further pretrained on ~40 k complete genomes'
- value
- The complete-genome checkpoints are initialised from the corresponding MAG checkpoint and further pretrained on about 40,000 complete genomes from NCBI RefSeq, so the MAG and complete-genome checkpoints are different configurations of the same architecture
Related records
- subject: Bacformer