Pipeline

DNABERT-2

Frozen DNABERT-2 mean-pooled last-hidden-state embeddings with task-specific logistic-regression or mini-batch k-means probe. Exact source checkpoint/evaluator revision is unreported.

16 evaluations · 16 metric rows

Overview

Frozen DNABERT-2 mean-pooled last-hidden-state embeddings with task-specific logistic-regression or mini-batch k-means probe. Exact source checkpoint/evaluator revision is unreported.

Consult the linked sources for architecture or protocol details. Missing evidence is not evidence of a missing capability.

Evaluations and results

16 evaluations · 16 metric rows. Different protocols are not a single leaderboard.

Filter evaluations

Tested model or methodProtocolDatasetEvaluation setupMetricEvidence origin

Applied filters: All linked evaluations

Exact evaluated configurations and original reported results
Tested configuration	Protocol and dataset	Finding	Evidence and details
Pipeline: DNABERT-2	Protocol: Gene-MTEB genomic embedding benchmark HMPD-disease: HMPD-disease genomic embedding evaluation Dataset subset: HMPD-disease Gene-MTEB data used in METAGENE-1 Table 3 (Gene-MTEB genomic embedding benchmark split)	0.480 accuracy fraction · higher Uncertainty: Not reported Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · source checked Methods, coverage and source DNABERT-2 on Gene-MTEB genomic embedding benchmark HMPD-disease: HMPD-disease genomic embedding evaluation Frozen model last-hidden-state embeddings, mean pooled. Logistic regression for eight classification tasks and mini-batch k-means for eight clustering tasks. Source paper Section 5.3. Current official gene-mteb implementation corroborates task identities but its commit is not proven identical to the paper evaluator. Exact dataset release, split hash, scored counts and evaluator commit are unreported in the paper. Do not substitute current mutable HF main for the original split. Aggregation: Not reported METAGENE-1 v1: Gene-MTEB complete component tasks · Table 3 (S5.T3), row HMPD-disease, column DNABERT-2; Section 5.3 paragraphs S5.SS3.p2.1 and S5.SS3.p3.1
Pipeline: DNABERT-2	Protocol: Gene-MTEB genomic embedding benchmark HMPD-sex: HMPD-sex genomic embedding evaluation Dataset subset: HMPD-sex Gene-MTEB data used in METAGENE-1 Table 3 (Gene-MTEB genomic embedding benchmark split)	0.366 accuracy fraction · higher Uncertainty: Not reported Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · source checked Methods, coverage and source DNABERT-2 on Gene-MTEB genomic embedding benchmark HMPD-sex: HMPD-sex genomic embedding evaluation Frozen model last-hidden-state embeddings, mean pooled. Logistic regression for eight classification tasks and mini-batch k-means for eight clustering tasks. Source paper Section 5.3. Current official gene-mteb implementation corroborates task identities but its commit is not proven identical to the paper evaluator. Exact dataset release, split hash, scored counts and evaluator commit are unreported in the paper. Do not substitute current mutable HF main for the original split. Aggregation: Not reported METAGENE-1 v1: Gene-MTEB complete component tasks · Table 3 (S5.T3), row HMPD-sex, column DNABERT-2; Section 5.3 paragraphs S5.SS3.p2.1 and S5.SS3.p3.1
Pipeline: DNABERT-2	Protocol: Gene-MTEB genomic embedding benchmark HMPD-single: HMPD-single genomic embedding evaluation Dataset subset: HMPD-single Gene-MTEB data used in METAGENE-1 Table 3 (Gene-MTEB genomic embedding benchmark split)	0.292 accuracy fraction · higher Uncertainty: Not reported Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · source checked Methods, coverage and source DNABERT-2 on Gene-MTEB genomic embedding benchmark HMPD-single: HMPD-single genomic embedding evaluation Frozen model last-hidden-state embeddings, mean pooled. Logistic regression for eight classification tasks and mini-batch k-means for eight clustering tasks. Source paper Section 5.3. Current official gene-mteb implementation corroborates task identities but its commit is not proven identical to the paper evaluator. Exact dataset release, split hash, scored counts and evaluator commit are unreported in the paper. Do not substitute current mutable HF main for the original split. Aggregation: Not reported METAGENE-1 v1: Gene-MTEB complete component tasks · Table 3 (S5.T3), row HMPD-single, column DNABERT-2; Section 5.3 paragraphs S5.SS3.p2.1 and S5.SS3.p3.1
Pipeline: DNABERT-2	Protocol: Gene-MTEB genomic embedding benchmark HMPD-source: HMPD-source genomic embedding evaluation Dataset subset: HMPD-source Gene-MTEB data used in METAGENE-1 Table 3 (Gene-MTEB genomic embedding benchmark split)	0.451 accuracy fraction · higher Uncertainty: Not reported Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · source checked Methods, coverage and source DNABERT-2 on Gene-MTEB genomic embedding benchmark HMPD-source: HMPD-source genomic embedding evaluation Frozen model last-hidden-state embeddings, mean pooled. Logistic regression for eight classification tasks and mini-batch k-means for eight clustering tasks. Source paper Section 5.3. Current official gene-mteb implementation corroborates task identities but its commit is not proven identical to the paper evaluator. Exact dataset release, split hash, scored counts and evaluator commit are unreported in the paper. Do not substitute current mutable HF main for the original split. Aggregation: Not reported METAGENE-1 v1: Gene-MTEB complete component tasks · Table 3 (S5.T3), row HMPD-source, column DNABERT-2; Section 5.3 paragraphs S5.SS3.p2.1 and S5.SS3.p3.1
Pipeline: DNABERT-2	Protocol: Gene-MTEB genomic embedding benchmark HMPR-p2p: HMPR-p2p genomic embedding evaluation Dataset subset: HMPR-p2p Gene-MTEB data used in METAGENE-1 Table 3 (Gene-MTEB genomic embedding benchmark split)	0.566 v_measure dimensionless · higher Uncertainty: Not reported Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · source checked Methods, coverage and source DNABERT-2 on Gene-MTEB genomic embedding benchmark HMPR-p2p: HMPR-p2p genomic embedding evaluation Frozen model last-hidden-state embeddings, mean pooled. Logistic regression for eight classification tasks and mini-batch k-means for eight clustering tasks. Source paper Section 5.3. Current official gene-mteb implementation corroborates task identities but its commit is not proven identical to the paper evaluator. Exact dataset release, split hash, scored counts and evaluator commit are unreported in the paper. Do not substitute current mutable HF main for the original split. Aggregation: Not reported METAGENE-1 v1: Gene-MTEB complete component tasks · Table 3 (S5.T3), row HMPR-p2p, column DNABERT-2; Section 5.3 paragraphs S5.SS3.p2.1 and S5.SS3.p3.1
Pipeline: DNABERT-2	Protocol: Gene-MTEB genomic embedding benchmark HMPR-s2s-align: HMPR-s2s-align genomic embedding evaluation Dataset subset: HMPR-s2s-align Gene-MTEB data used in METAGENE-1 Table 3 (Gene-MTEB genomic embedding benchmark split)	0.127 v_measure dimensionless · higher Uncertainty: Not reported Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · source checked Methods, coverage and source DNABERT-2 on Gene-MTEB genomic embedding benchmark HMPR-s2s-align: HMPR-s2s-align genomic embedding evaluation Frozen model last-hidden-state embeddings, mean pooled. Logistic regression for eight classification tasks and mini-batch k-means for eight clustering tasks. Source paper Section 5.3. Current official gene-mteb implementation corroborates task identities but its commit is not proven identical to the paper evaluator. Exact dataset release, split hash, scored counts and evaluator commit are unreported in the paper. Do not substitute current mutable HF main for the original split. Aggregation: Not reported METAGENE-1 v1: Gene-MTEB complete component tasks · Table 3 (S5.T3), row HMPR-s2s-align, column DNABERT-2; Section 5.3 paragraphs S5.SS3.p2.1 and S5.SS3.p3.1
Pipeline: DNABERT-2	Protocol: Gene-MTEB genomic embedding benchmark HMPR-s2s-small: HMPR-s2s-small genomic embedding evaluation Dataset subset: HMPR-s2s-small Gene-MTEB data used in METAGENE-1 Table 3 (Gene-MTEB genomic embedding benchmark split)	0.419 v_measure dimensionless · higher Uncertainty: Not reported Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · source checked Methods, coverage and source DNABERT-2 on Gene-MTEB genomic embedding benchmark HMPR-s2s-small: HMPR-s2s-small genomic embedding evaluation Frozen model last-hidden-state embeddings, mean pooled. Logistic regression for eight classification tasks and mini-batch k-means for eight clustering tasks. Source paper Section 5.3. Current official gene-mteb implementation corroborates task identities but its commit is not proven identical to the paper evaluator. Exact dataset release, split hash, scored counts and evaluator commit are unreported in the paper. Do not substitute current mutable HF main for the original split. Aggregation: Not reported METAGENE-1 v1: Gene-MTEB complete component tasks · Table 3 (S5.T3), row HMPR-s2s-small, column DNABERT-2; Section 5.3 paragraphs S5.SS3.p2.1 and S5.SS3.p3.1
Pipeline: DNABERT-2	Protocol: Gene-MTEB genomic embedding benchmark HMPR-s2s-tiny: HMPR-s2s-tiny genomic embedding evaluation Dataset subset: HMPR-s2s-tiny Gene-MTEB data used in METAGENE-1 Table 3 (Gene-MTEB genomic embedding benchmark split)	0.274 v_measure dimensionless · higher Uncertainty: Not reported Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · source checked Methods, coverage and source DNABERT-2 on Gene-MTEB genomic embedding benchmark HMPR-s2s-tiny: HMPR-s2s-tiny genomic embedding evaluation Frozen model last-hidden-state embeddings, mean pooled. Logistic regression for eight classification tasks and mini-batch k-means for eight clustering tasks. Source paper Section 5.3. Current official gene-mteb implementation corroborates task identities but its commit is not proven identical to the paper evaluator. Exact dataset release, split hash, scored counts and evaluator commit are unreported in the paper. Do not substitute current mutable HF main for the original split. Aggregation: Not reported METAGENE-1 v1: Gene-MTEB complete component tasks · Table 3 (S5.T3), row HMPR-s2s-tiny, column DNABERT-2; Section 5.3 paragraphs S5.SS3.p2.1 and S5.SS3.p3.1
Pipeline: DNABERT-2	Protocol: Gene-MTEB genomic embedding benchmark Human-Virus-1: Human-Virus-1 genomic embedding evaluation Dataset subset: Human-Virus-1 Gene-MTEB data used in METAGENE-1 Table 3 (Gene-MTEB genomic embedding benchmark split)	0.594 accuracy fraction · higher Uncertainty: Not reported Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · source checked Methods, coverage and source DNABERT-2 on Gene-MTEB genomic embedding benchmark Human-Virus-1: Human-Virus-1 genomic embedding evaluation Frozen model last-hidden-state embeddings, mean pooled. Logistic regression for eight classification tasks and mini-batch k-means for eight clustering tasks. Source paper Section 5.3. Current official gene-mteb implementation corroborates task identities but its commit is not proven identical to the paper evaluator. Exact dataset release, split hash, scored counts and evaluator commit are unreported in the paper. Do not substitute current mutable HF main for the original split. Aggregation: Not reported METAGENE-1 v1: Gene-MTEB complete component tasks · Table 3 (S5.T3), row Human-Virus-1, column DNABERT-2; Section 5.3 paragraphs S5.SS3.p2.1 and S5.SS3.p3.1
Pipeline: DNABERT-2	Protocol: Gene-MTEB genomic embedding benchmark Human-Virus-2: Human-Virus-2 genomic embedding evaluation Dataset subset: Human-Virus-2 Gene-MTEB data used in METAGENE-1 Table 3 (Gene-MTEB genomic embedding benchmark split)	0.507 accuracy fraction · higher Uncertainty: Not reported Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · source checked Methods, coverage and source DNABERT-2 on Gene-MTEB genomic embedding benchmark Human-Virus-2: Human-Virus-2 genomic embedding evaluation Frozen model last-hidden-state embeddings, mean pooled. Logistic regression for eight classification tasks and mini-batch k-means for eight clustering tasks. Source paper Section 5.3. Current official gene-mteb implementation corroborates task identities but its commit is not proven identical to the paper evaluator. Exact dataset release, split hash, scored counts and evaluator commit are unreported in the paper. Do not substitute current mutable HF main for the original split. Aggregation: Not reported METAGENE-1 v1: Gene-MTEB complete component tasks · Table 3 (S5.T3), row Human-Virus-2, column DNABERT-2; Section 5.3 paragraphs S5.SS3.p2.1 and S5.SS3.p3.1
Pipeline: DNABERT-2	Protocol: Gene-MTEB genomic embedding benchmark Human-Virus-3: Human-Virus-3 genomic embedding evaluation Dataset subset: Human-Virus-3 Gene-MTEB data used in METAGENE-1 Table 3 (Gene-MTEB genomic embedding benchmark split)	0.606 accuracy fraction · higher Uncertainty: Not reported Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · source checked Methods, coverage and source DNABERT-2 on Gene-MTEB genomic embedding benchmark Human-Virus-3: Human-Virus-3 genomic embedding evaluation Frozen model last-hidden-state embeddings, mean pooled. Logistic regression for eight classification tasks and mini-batch k-means for eight clustering tasks. Source paper Section 5.3. Current official gene-mteb implementation corroborates task identities but its commit is not proven identical to the paper evaluator. Exact dataset release, split hash, scored counts and evaluator commit are unreported in the paper. Do not substitute current mutable HF main for the original split. Aggregation: Not reported METAGENE-1 v1: Gene-MTEB complete component tasks · Table 3 (S5.T3), row Human-Virus-3, column DNABERT-2; Section 5.3 paragraphs S5.SS3.p2.1 and S5.SS3.p3.1
Pipeline: DNABERT-2	Protocol: Gene-MTEB genomic embedding benchmark Human-Virus-4: Human-Virus-4 genomic embedding evaluation Dataset subset: Human-Virus-4 Gene-MTEB data used in METAGENE-1 Table 3 (Gene-MTEB genomic embedding benchmark split)	0.550 accuracy fraction · higher Uncertainty: Not reported Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · source checked Methods, coverage and source DNABERT-2 on Gene-MTEB genomic embedding benchmark Human-Virus-4: Human-Virus-4 genomic embedding evaluation Frozen model last-hidden-state embeddings, mean pooled. Logistic regression for eight classification tasks and mini-batch k-means for eight clustering tasks. Source paper Section 5.3. Current official gene-mteb implementation corroborates task identities but its commit is not proven identical to the paper evaluator. Exact dataset release, split hash, scored counts and evaluator commit are unreported in the paper. Do not substitute current mutable HF main for the original split. Aggregation: Not reported METAGENE-1 v1: Gene-MTEB complete component tasks · Table 3 (S5.T3), row Human-Virus-4, column DNABERT-2; Section 5.3 paragraphs S5.SS3.p2.1 and S5.SS3.p3.1
Pipeline: DNABERT-2	Protocol: Gene-MTEB genomic embedding benchmark HVR-p2p: HVR-p2p genomic embedding evaluation Dataset subset: HVR-p2p Gene-MTEB data used in METAGENE-1 Table 3 (Gene-MTEB genomic embedding benchmark split)	0.548 v_measure dimensionless · higher Uncertainty: Not reported Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · source checked Methods, coverage and source DNABERT-2 on Gene-MTEB genomic embedding benchmark HVR-p2p: HVR-p2p genomic embedding evaluation Frozen model last-hidden-state embeddings, mean pooled. Logistic regression for eight classification tasks and mini-batch k-means for eight clustering tasks. Source paper Section 5.3. Current official gene-mteb implementation corroborates task identities but its commit is not proven identical to the paper evaluator. Exact dataset release, split hash, scored counts and evaluator commit are unreported in the paper. Do not substitute current mutable HF main for the original split. Aggregation: Not reported METAGENE-1 v1: Gene-MTEB complete component tasks · Table 3 (S5.T3), row HVR-p2p, column DNABERT-2; Section 5.3 paragraphs S5.SS3.p2.1 and S5.SS3.p3.1
Pipeline: DNABERT-2	Protocol: Gene-MTEB genomic embedding benchmark HVR-s2s-align: HVR-s2s-align genomic embedding evaluation Dataset subset: HVR-s2s-align Gene-MTEB data used in METAGENE-1 Table 3 (Gene-MTEB genomic embedding benchmark split)	0.243 v_measure dimensionless · higher Uncertainty: Not reported Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · source checked Methods, coverage and source DNABERT-2 on Gene-MTEB genomic embedding benchmark HVR-s2s-align: HVR-s2s-align genomic embedding evaluation Frozen model last-hidden-state embeddings, mean pooled. Logistic regression for eight classification tasks and mini-batch k-means for eight clustering tasks. Source paper Section 5.3. Current official gene-mteb implementation corroborates task identities but its commit is not proven identical to the paper evaluator. Exact dataset release, split hash, scored counts and evaluator commit are unreported in the paper. Do not substitute current mutable HF main for the original split. Aggregation: Not reported METAGENE-1 v1: Gene-MTEB complete component tasks · Table 3 (S5.T3), row HVR-s2s-align, column DNABERT-2; Section 5.3 paragraphs S5.SS3.p2.1 and S5.SS3.p3.1
Pipeline: DNABERT-2	Protocol: Gene-MTEB genomic embedding benchmark HVR-s2s-small: HVR-s2s-small genomic embedding evaluation Dataset subset: HVR-s2s-small Gene-MTEB data used in METAGENE-1 Table 3 (Gene-MTEB genomic embedding benchmark split)	0.373 v_measure dimensionless · higher Uncertainty: Not reported Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · source checked Methods, coverage and source DNABERT-2 on Gene-MTEB genomic embedding benchmark HVR-s2s-small: HVR-s2s-small genomic embedding evaluation Frozen model last-hidden-state embeddings, mean pooled. Logistic regression for eight classification tasks and mini-batch k-means for eight clustering tasks. Source paper Section 5.3. Current official gene-mteb implementation corroborates task identities but its commit is not proven identical to the paper evaluator. Exact dataset release, split hash, scored counts and evaluator commit are unreported in the paper. Do not substitute current mutable HF main for the original split. Aggregation: Not reported METAGENE-1 v1: Gene-MTEB complete component tasks · Table 3 (S5.T3), row HVR-s2s-small, column DNABERT-2; Section 5.3 paragraphs S5.SS3.p2.1 and S5.SS3.p3.1
Pipeline: DNABERT-2	Protocol: Gene-MTEB genomic embedding benchmark HVR-s2s-tiny: HVR-s2s-tiny genomic embedding evaluation Dataset subset: HVR-s2s-tiny Gene-MTEB data used in METAGENE-1 Table 3 (Gene-MTEB genomic embedding benchmark split)	0.753 v_measure dimensionless · higher Uncertainty: Not reported Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · source checked Methods, coverage and source DNABERT-2 on Gene-MTEB genomic embedding benchmark HVR-s2s-tiny: HVR-s2s-tiny genomic embedding evaluation Frozen model last-hidden-state embeddings, mean pooled. Logistic regression for eight classification tasks and mini-batch k-means for eight clustering tasks. Source paper Section 5.3. Current official gene-mteb implementation corroborates task identities but its commit is not proven identical to the paper evaluator. Exact dataset release, split hash, scored counts and evaluator commit are unreported in the paper. Do not substitute current mutable HF main for the original split. Aggregation: Not reported METAGENE-1 v1: Gene-MTEB complete component tasks · Table 3 (S5.T3), row HVR-s2s-tiny, column DNABERT-2; Section 5.3 paragraphs S5.SS3.p2.1 and S5.SS3.p3.1

Source checking is not independent reproduction. Release 2026-09-23-2b89723c6dd9.

Use this model

How it works, versions and access

Underlying model: DNABERT-2. Results on this page belong to this pipeline and its evaluated settings.

Strengths, limitations and unresolved questions

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.

1 evidence row matching the loaded filters

Claims, original sources and review scope · Release 2026-09-23-2b89723c6dd9
Property and statement	Original source and location	Review and provenance
Relationship: uses model discovery-model-dnabert-2 Individual claims	METAGENE-1 v1: Gene-MTEB complete component tasks Original source ↗ Section 5.3; Table 3 DNABERT-2 column Version: arXiv:2501.02045v1 Retrieved: 2026-09-23T11:22:00.101850+00:00	source checked automated source review · 2026-09-23 Audit details Source-backed evaluated identity only; no independent reproduction. Field: `links:uses_model:discovery-model-dnabert-2` Claim: metagene-gene-mteb-method-dnabert-2-discovery-model-dnabert-2-identity-claim Source artifact SHA-256: `a426c77d137138cfc33410a1e54a1cce521007bdc9e3176f4774d0105424a7fe` Hash scope: Hash scope not separately documented; inspect source record Inspected artifact

All evidence (CSV)All evidence (JSONL)How to read the evidence

Sources and history

View linked audit checks and correction history

Release 2026-09-23-2b89723c6dd9 · Record review: source checked

1 source records and release history

METAGENE-1 v1: Gene-MTEB complete component tasks · Original source · arXiv:2501.02045v1

Download this release

Technical metadata and extraction receipts

Stable ID: metagene-gene-mteb-method-dnabert-2

areas: dna-genomes
source locator: Section 5.3; Table 3 DNABERT-2 column
missing metadata: checkpoint revision: unreported; parameters: unextracted

Related records

Suggest a correction