Result

Accelerator time (Pretraining, total) of DNABERT 110M pretraining (HyenaDNA Table A.2) on Pretraining corpora of the models in HyenaDNA Table A.2

12,000 Accelerator time (Pretraining, total)

Evidence concern: excluded from comparisons

Table A.2 GPU-hours for the DNABERT and Nucleotide Transformer columns do not equal the GPUs × wall clock printed in the same table: 8 GPUs × 25 days is 4,800 GPU-hours against 12,000 printed, and 128 GPUs × 28 days is 86,016 GPU-hours against 215,000 printed, both about 2.5 times higher. The HyenaDNA columns agree (1 GPU × 80 min, about 1.3 GPU-hours). The Nucleotide Transformer article (Methods 'Hardware') also gives 128 A100 GPUs for 28 days. The table cites no source for these columns. Values are kept as printed and should not be compared until the discrepancy is resolved.HyenaDNA: Long-Range Genomic Sequence Modeling at Single Nucleotide Resolution, arXiv v2 · Table A.2 (page 17), rows 'GPUs', 'Wall clock' and 'GPU-hrs', columns 'DNABERT' and 'Nucleotide Transformer'

Methods

Tested configuration: DNABERT 110M pretraining (HyenaDNA Table A.2)
Protocol: Pretraining GPUs and runtime for short-range models (HyenaDNA Table A.2)
Dataset: Pretraining corpora of the models in HyenaDNA Table A.2
Procedure: Pretraining GPUs and runtime for short-range models (HyenaDNA Table A.2)
Evaluation: DNABERT 110M pretraining compute (HyenaDNA Table A.2)
Coverage: Not reported scored / Not reported eligible
Uncertainty: Not reported by the source
Evidence: Result quoted from another source · source checkedHyenaDNA: Long-Range Genomic Sequence Modeling at Single Nucleotide Resolution, arXiv v2 · Table A.2 (page 17), row 'GPU-hrs', column 1 of 4 ('DNABERT', 110M)

A source-checked result verifies the numerical transcription, not every model or protocol detail. Evaluation metadata: source checked. Source checked does not mean independently reproduced.

Reproduction

Split: Not applicable
Adaptation: Pretraining from scratch
Scoring implementation: Not reported

No execution recipe has been verified for this exact configuration and evaluation. A benchmark's general instructions may use different inputs, splits or model settings.

Reproducing this published result requires matching its model configuration, data, split and scorer. Source checking or a successful smoke test does not establish score reproduction.

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 2026-10-22-1b7cf027e4f6
Property and statement	Original source and location	Review and provenance
Reported result 12,000 Individual claims	HyenaDNA: Long-Range Genomic Sequence Modeling at Single Nucleotide Resolution, arXiv v2 Original source ↗ Table A.2 (page 17), row 'GPU-hrs', column 1 of 4 ('DNABERT', 110M) Version: arXiv 2306.15794v2 Retrieved: 2026-10-14T12:48:26Z	source checked ["source-hash-verification","pdf-text-parse","independent-cell-check"] · 2026-10-14 paper compilation Source has a recorded evidence concern. Consult its source page before using the claim. Audit details Value matches. paper_compilation accepted: the table gives no source for this row. GPU-hours are about 2.5 times GPUs × wall clock in the same table; see the source concern added by the review. Transcribed, not reproduced. Field: `attributes.printed_value` Source artifact SHA-256: `2f471549352ff571ac55faa80c8da453f9bf1ff5b01b6b87fa893457828e798c` Hash scope: Hash scope not separately documented; inspect source record Inspected artifact Extraction artifact SHA-256: `2f471549352ff571ac55faa80c8da453f9bf1ff5b01b6b87fa893457828e798c` Extraction artifact

How to read the evidence

Sources and history

Release 2026-10-22-1b7cf027e4f6 · Record review: source checked

1 source record and release history

HyenaDNA: Long-Range Genomic Sequence Modeling at Single Nucleotide Resolution, arXiv v2 · Original source · arXiv 2306.15794v2

Download this release (gzip)

Technical metadata and extraction receipts

Stable ID: gfm-guide-20261014-result-hyenadna-table-a2-dnabert-110m-gpu-hours

metric: accelerator-time
metric direction: lower
unit: hour
printed value: 12,000
numeric value: 12000
source locator: Table A.2 (page 17), row 'GPU-hrs', column 1 of 4 ('DNABERT', 110M)
missing metadata: uncertainty: reason: unreported
review: method: source-hash-verification; pdf-text-parse; independent-cell-check; method note: Table A.2 (page 17) parsed by the reviewer's script (review/check_gfm_guide.py) from pdftotext -layout; caption, Params, GPUs, Wall clock and GPU-hrs rows asserted; PDF SHA-256 matches.; reviewer: claude; reviewer note: Independent Claude (Opus 5.5) review agent for PR #150; did not extract this batch; no human review claimed; date: 2026-10-14; artifact sha256: 2f471549352ff571ac55faa80c8da453f9bf1ff5b01b6b87fa893457828e798c; retrieval url: https://arxiv.org/pdf/2306.15794v2; note: Value matches. paper_compilation accepted: the table gives no source for this row. GPU-hours are about 2.5 times GPUs × wall clock in the same table; see the source concern added by the review. Transcribed, not reproduced.
metric qualifier: Pretraining, total
unit detail: GPU-hours on 8 × NVIDIA RTX 2080 Ti
limitations: Printed GPU-hours are about 2.5 times the GPUs × wall clock printed in the same column; see the source concern. Do not compare without resolving it.

Related records

evaluation: DNABERT 110M pretraining compute (HyenaDNA Table A.2)

Suggest a correction