Model type
Single-cell transformer; this record is the paper-specific evaluated configuration.
Configuration
Geneformer
Geneformer is an expression-pretrained comparator in the Cell2Sentence label-prediction study.
Sources
Cell2Sentence: Teaching Large Language Models the Language of Biology · Discussion (paragraph 1); Methods (paragraph 1)How it worksEvaluated procedure (conceptual)
Conceptual input–method–output guide. Check the procedure text and linked evaluation for fitted components, additional inputs and exact settings.
Sources
Cell2Sentence: Teaching Large Language Models the Language of Biology · Methods/Data transformation (paragraph 7); Methods/Data transformation (paragraph 6)At a glance
Biological inputs
Rank-encoded transcriptomic profiles
Sources
Cell2Sentence: Teaching Large Language Models the Language of Biology · Methods/Data transformation (paragraph 7); Methods/Data transformation (paragraph 6)Outputs
Cell/sample representations and label predictions
Sources
Cell2Sentence: Teaching Large Language Models the Language of Biology · Experiments/Fine-Tuning Datasets (paragraph 5); Inference Details (paragraph 1)limited source coverage · Automated source review, 2026-09-16. All specifications and missing details
Evaluations and results
Release 2026-09-17-d277315f7d76 · 1 evaluation · 1 metric row. Different protocols are not a single leaderboard.
| Metric and finding | Coverage and uncertainty | Evidence |
|---|---|---|
| Geneformer: Combinatorial cell-label classification Partial-credit labels including cell type, perturbation, and dose. Independent external evaluation · Evaluation metadata: needs review | ||
| 0.419 Partial-label accuracy Unit: unitless · Direction: unknown | Uncertainty: ± 0.0153 Scored: Not reported · Eligible: Not reported | source checkedCell2Sentence: Teaching Large Language Models the Language of Biology · Table 3, Partial label / Geneformer row, L1000 Acc column Source checking is not independent reproduction. |
How it works
How the evaluated method works
Rank-encoded gene-expression input passes through a pretrained transformer; the study evaluates its representation on label-prediction tasks.
Sources
Cell2Sentence: Teaching Large Language Models the Language of Biology · Methods/Data transformation (paragraph 7); Methods/Data transformation (paragraph 6)Underlying method and version boundaries
Geneformer ranks genes by expression scaled against its pretraining corpus, then uses a transformer encoder with a masked-gene objective. V1 and V2 have different corpora, vocabularies, sizes and context limits, so a historical paper name is not replaced with today’s default checkpoint.
Sources
huggingface.co/ctheodoris/Geneformer README.md · README.md; introduction, model description, pretrained-model and usage sections at pinned revisionWhat was evaluated
The linked evaluation record identifies Geneformer: Combinatorial cell-label classification. Its dataset, split, adaptation and evidence origin remain attached to the reported results.
Sources
Cell2Sentence: Teaching Large Language Models the Language of Biology · The named evaluation’s methods and comparison table; exact preserved evaluation IDs: evaluation-lit-028Strengths and limitations
Strengths and considerations
- Supplies a single-cell-pretrained reference for evaluating the C2S representation.
Sources
Cell2Sentence: Teaching Large Language Models the Language of Biology · Methods/Data transformation (paragraph 8); Generated abstracts/Examples of generated abstracts. (paragraph 1)
Limitations and conditions
- Bulk L1000 and GTEx evaluations are distribution shifts from single-cell pretraining; the C2S method’s architecture must not be assigned to this comparator.
Sources
Cell2Sentence: Teaching Large Language Models the Language of Biology · Experiments/Experiment 2: cell label prediction/Methodology: (paragraph 1); Experimental Details/Evaluation Datasets (paragraph 1)
Profile review details
Primary full text and the available official implementation/model documentation were inspected. Explanatory claims are source-backed; unresolved exact-configuration metadata is labelled explicitly. This is automated review, not a human review or independent benchmark reproduction.
Stable record: reported-model-06816ce9073144Specifications
Inputs, training, access and other details
Explanatory profile: limited source coverage · Automated source review, 2026-09-16. Review applies to the cited claims; unresolved fields are listed below. Numerical results retain their own review status.
| Property | Description and evidence |
|---|---|
| Model type | Single-cell transformer; this record is the paper-specific evaluated configuration.Sourceshuggingface.co/ctheodoris/Geneformer README.md · README.md model description |
| Architecture / procedure | Rank-encoded gene-expression input passes through a pretrained transformer; the study evaluates its representation on label-prediction tasks.SourcesCell2Sentence: Teaching Large Language Models the Language of Biology · Methods/Data transformation (paragraph 7); Methods/Data transformation (paragraph 6) |
| Biological inputs | Rank-encoded transcriptomic profilesSourcesCell2Sentence: Teaching Large Language Models the Language of Biology · Methods/Data transformation (paragraph 7); Methods/Data transformation (paragraph 6) |
| Outputs | Cell/sample representations and label predictionsSourcesCell2Sentence: Teaching Large Language Models the Language of Biology · Experiments/Fine-Tuning Datasets (paragraph 5); Inference Details (paragraph 1) |
| Parameters | 46,107,089 parameters for the model checkpoint used in the Cell2Sentence L1000 experiment; this is not a claim about every checkpoint in the family.SourcesCell2Sentence: Teaching Large Language Models the Language of Biology · Table 5, Comparison of required compute on the L1000 dataset; # Parameters column and caption |
| Known versions / configuration | Geneformer is the comparison-table label; that label does not specify an immutable weight revision. · Not reported in inspected sourcesSourcesCell2Sentence: Teaching Large Language Models the Language of Biology · Model identification in the comparison table and corresponding Methods; immutable checkpoint revision is not supplied by the table label. |
| Training data / fitting | The reported comparator transfers its pretrained single-cell representation to the study’s label-prediction tasks, including out-of-distribution L1000 and GTEx bulk data and stimulated human PBMCs. Its precise pretraining checkpoint is not pinned by the comparison passage.SourcesCell2Sentence: Teaching Large Language Models the Language of Biology · Experiments / Experiment 2: cell label prediction / Methodology |
| Context limits | A maximum input/context length for this exact evaluated configuration is not established by the inspected sources. · Not reported in inspected sourcesSources (2)Cell2Sentence: Teaching Large Language Models the Language of Biology; huggingface.co/ctheodoris/Geneformer README.md · Background and Related Work/Multimodal Training and Cross-Modality Encoding; Methods; Methods/Data transformation; Methods/Transformation robustness; Methods/Tasks; Experiments/Experiment 2: cell label prediction/Methodology:; Experiments/Experiment 3: abstract summary generation/Methodology:; Method Details; inspected for explicit maximum input length (dataset lengths and family-wide limits are not substituted); README.md at pinned repository revision |
| Access | Official upstream implementation and usage documentation: https://huggingface.co/ctheodoris/Geneformer/blob/1f7fbae4e469a5f4f1af8c111a529cfe1b3829f5/README.md. This pinned documentation revision is not automatically the evaluated weight revision.Sourceshuggingface.co/ctheodoris/Geneformer README.md · README.md; installation, model download and usage instructions |
| Code licence | No explicit code licence was established from the paper’s availability statement and inspected repository-root documentation. · Not reported in inspected sourcesSourceshuggingface.co/ctheodoris/Geneformer README.md · README.md and repository-root licence-file search |
| Weights licence | Apache 2.0 is declared in the official Geneformer model-card metadata; the exact historical configuration still needs its checkpoint identity.Sourceshuggingface.co/ctheodoris/Geneformer README.md · README.md front matter, license field; model-version list |
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 |
|---|---|---|
| Diagram caption Conceptual input–method–output guide. Check the procedure text and linked evaluation for fitted components, additional inputs and exact settings. Individual claims | Cell2Sentence: Teaching Large Language Models the Language of Biology Methods/Data transformation (paragraph 7); Methods/Data transformation (paragraph 6) Version: preprint archived 2024-10-29 | source checked automated source review · 2026-09-16 Audit detailsPrimary full text and the available official implementation/model documentation were inspected. Explanatory claims are source-backed; unresolved exact-configuration metadata is labelled explicitly. This is automated review, not a human review or independent benchmark reproduction. Field: Source artifact SHA-256: Hash scope: Hash scope not separately documented; inspect source record |
| Diagram steps ["Rank-encoded transcriptomic profiles","Geneformer","Cell/sample representations and label predictions"] Individual claims | Cell2Sentence: Teaching Large Language Models the Language of Biology Methods/Data transformation (paragraph 7); Methods/Data transformation (paragraph 6) Version: preprint archived 2024-10-29 | source checked automated source review · 2026-09-16 Audit detailsPrimary full text and the available official implementation/model documentation were inspected. Explanatory claims are source-backed; unresolved exact-configuration metadata is labelled explicitly. This is automated review, not a human review or independent benchmark reproduction. Field: Source artifact SHA-256: Hash scope: Hash scope not separately documented; inspect source record |
| Diagram title Evaluated procedure (conceptual) Individual claims | Cell2Sentence: Teaching Large Language Models the Language of Biology Methods/Data transformation (paragraph 7); Methods/Data transformation (paragraph 6) Version: preprint archived 2024-10-29 | source checked automated source review · 2026-09-16 Audit detailsPrimary full text and the available official implementation/model documentation were inspected. Explanatory claims are source-backed; unresolved exact-configuration metadata is labelled explicitly. This is automated review, not a human review or independent benchmark reproduction. Field: Source artifact SHA-256: Hash scope: Hash scope not separately documented; inspect source record |
| Model type Single-cell transformer; this record is the paper-specific evaluated configuration. Individual claims | huggingface.co/ctheodoris/Geneformer README.md README.md model description Version: 1f7fbae4e469a5f4f1af8c111a529cfe1b3829f5 | source checked automated source review · 2026-09-16 Audit detailsPrimary full text and the available official implementation/model documentation were inspected. Explanatory claims are source-backed; unresolved exact-configuration metadata is labelled explicitly. This is automated review, not a human review or independent benchmark reproduction. Field: Source artifact SHA-256: Hash scope: Hash scope not separately documented; inspect source record |
| Architecture / procedure Rank-encoded gene-expression input passes through a pretrained transformer; the study evaluates its representation on label-prediction tasks. Individual claims | Cell2Sentence: Teaching Large Language Models the Language of Biology Methods/Data transformation (paragraph 7); Methods/Data transformation (paragraph 6) Version: preprint archived 2024-10-29 | source checked automated source review · 2026-09-16 Audit detailsPrimary full text and the available official implementation/model documentation were inspected. Explanatory claims are source-backed; unresolved exact-configuration metadata is labelled explicitly. This is automated review, not a human review or independent benchmark reproduction. Field: Source artifact SHA-256: Hash scope: Hash scope not separately documented; inspect source record |
| Weights licence Apache 2.0 is declared in the official Geneformer model-card metadata; the exact historical configuration still needs its checkpoint identity. Individual claims | huggingface.co/ctheodoris/Geneformer README.md README.md front matter, license field; model-version list Version: 1f7fbae4e469a5f4f1af8c111a529cfe1b3829f5 | source checked automated source review · 2026-09-16 Audit detailsPrimary full text and the available official implementation/model documentation were inspected. Explanatory claims are source-backed; unresolved exact-configuration metadata is labelled explicitly. This is automated review, not a human review or independent benchmark reproduction. Field: Source artifact SHA-256: Hash scope: Hash scope not separately documented; inspect source record |
| Biological inputs Rank-encoded transcriptomic profiles Individual claims | Cell2Sentence: Teaching Large Language Models the Language of Biology Methods/Data transformation (paragraph 7); Methods/Data transformation (paragraph 6) Version: preprint archived 2024-10-29 | source checked automated source review · 2026-09-16 Audit detailsPrimary full text and the available official implementation/model documentation were inspected. Explanatory claims are source-backed; unresolved exact-configuration metadata is labelled explicitly. This is automated review, not a human review or independent benchmark reproduction. Field: Source artifact SHA-256: Hash scope: Hash scope not separately documented; inspect source record |
| Outputs Cell/sample representations and label predictions Individual claims | Cell2Sentence: Teaching Large Language Models the Language of Biology Experiments/Fine-Tuning Datasets (paragraph 5); Inference Details (paragraph 1) Version: preprint archived 2024-10-29 | source checked automated source review · 2026-09-16 Audit detailsPrimary full text and the available official implementation/model documentation were inspected. Explanatory claims are source-backed; unresolved exact-configuration metadata is labelled explicitly. This is automated review, not a human review or independent benchmark reproduction. Field: Source artifact SHA-256: Hash scope: Hash scope not separately documented; inspect source record |
| Parameters 46,107,089 parameters for the model checkpoint used in the Cell2Sentence L1000 experiment; this is not a claim about every checkpoint in the family. Individual claims | Cell2Sentence: Teaching Large Language Models the Language of Biology Table 5, Comparison of required compute on the L1000 dataset; # Parameters column and caption Version: preprint archived 2024-10-29 | source checked automated source review · 2026-09-16 Audit detailsPrimary full text and the available official implementation/model documentation were inspected. Explanatory claims are source-backed; unresolved exact-configuration metadata is labelled explicitly. This is automated review, not a human review or independent benchmark reproduction. Field: Source artifact SHA-256: Hash scope: Hash scope not separately documented; inspect source record |
| Known versions / configuration Geneformer is the comparison-table label; that label does not specify an immutable weight revision. Individual claims | Cell2Sentence: Teaching Large Language Models the Language of Biology Model identification in the comparison table and corresponding Methods; immutable checkpoint revision is not supplied by the table label. Version: preprint archived 2024-10-29 | unreported automated source review · 2026-09-16 Audit detailsPrimary full text and the available official implementation/model documentation were inspected. Explanatory claims are source-backed; unresolved exact-configuration metadata is labelled explicitly. This is automated review, not a human review or independent benchmark reproduction. Field: Source artifact SHA-256: Hash scope: Hash scope not separately documented; inspect source record |
Sources and history
Release 2026-09-17-d277315f7d76 · Record review: needs review
2 source records and release history
- Cell2Sentence: Teaching Large Language Models the Language of Biology · Original source · preprint archived 2024-10-29
- huggingface.co/ctheodoris/Geneformer README.md · Original source · 1f7fbae4e469a5f4f1af8c111a529cfe1b3829f5
Technical metadata and extraction receipts
Stable ID: reported-model-06816ce9073144
- areas
- cells-tissues
- entity level
- method
- version
- Not reported
- reported name
- Geneformer
- historical missing metadata
- version: not_reported_in_legacy_extract; checkpoint revision: not_reported_in_legacy_extract; training data: not_reported_in_legacy_extract; licence: not_reported_in_legacy_extract
- metadata review scope
- historical_missing_metadata preserves the original discovery state. Current descriptive evidence and missingness are recorded in profile.facts; numerical-result review is separate.
- legacy kinds
- model
- entity classification
- review date: 2026-09-17; rationale: This source-scoped entry preserves the method/configuration actually named in an evaluation. It is neither a global family identity nor proof of an immutable checkpoint; the linked evaluation retains adaptation, fitting and scoring details.; source ids: cell2sentence-2024; evidence-reported-base-geneformer-readme-md; source locator: Methods/Data transformation (paragraph 7); Methods/Data transformation (paragraph 6) | README.md model description | Discussion (paragraph 1); Methods (paragraph 1); ambiguities: Configuration means the source-labelled evaluated identity. It does not establish missing checkpoint hashes, default settings or equivalence to same-named records in other papers.