rewire.it
Configuration

ESM-2 8M frozen residue-mean embeddings + fixed ridge (Rewire) (FLIP2 Rhomax by_wild_type)

ESM-2 is a family of protein sequence encoders that produce representations for downstream protein analyses.

Sources (3)facebookresearch/esm: README.md; facebook/esm2_t33_650M_UR50D: README.md; facebook/esm2_t33_650M_UR50D: config.json · ESM README: Pre-trained Models and Main models; official facebook/esm2_t33_650M_UR50D README licence metadata and config.json

1 evaluation · 2 metric rows

How it worksESM-2 workflow
ESM-2 workflow1. Protein sequence. Then: 2. Transformer layers. Then: 3. Residue embeddings. Then: 4. Specified downstream analysisESM-2 workflow1. Protein sequence. Then: 2. Transformer layers. Then: 3. Residue embeddings. Then: 4. Specified downstream analysisESM-2 workflow1. Protein sequence. Then: 2. Transformer layers. Then: 3. Residue embeddings. Then: 4. Specified downstream analysis

Conceptual summary of the documented data flow; optional inputs and configured downstream stages must be reported for a reproducible evaluation.

Sources (3)facebookresearch/esm: README.md; facebook/esm2_t33_650M_UR50D: README.md; facebook/esm2_t33_650M_UR50D: config.json · ESM README: Pre-trained Models and Main models; official facebook/esm2_t33_650M_UR50D README licence metadata and config.json

Overview

Source reviewed · Automated source review, 2026-09-16. All specifications and missing details

Evaluations and results

1 evaluation · 2 metric rows. Different protocols are not a single leaderboard.

Filter evaluations

Applied filters: All linked evaluations

Exact evaluated configurations and original reported results
Tested configurationProtocol and datasetFindingEvidence and details
Configuration: ESM-2 8M frozen residue-mean embeddings + fixed ridge (Rewire) (FLIP2 Rhomax by_wild_type)Protocol: FLIP2 Rhomax by_wild_type
Dataset subset: FLIP2 Rhomax by_wild_type test subset
0.8964799835636942 ndcg
dimensionless · higher

Uncertainty: Not reported

Coverage: 184/184

Rewire evaluation · source checked
Methods, coverage and source

ESM-2 8M frozen residue-mean embeddings + fixed ridge (Rewire) on FLIP2 Rhomax by_wild_type

FLIP2 Rhomax by_wild_type

Aggregation: Not reported

ESM-2 8M frozen residue-mean embeddings + fixed ridge (Rewire): report; ESM-2 8M frozen residue-mean embeddings + fixed ridge (Rewire): audit · metrics.ndcg
Configuration: ESM-2 8M frozen residue-mean embeddings + fixed ridge (Rewire) (FLIP2 Rhomax by_wild_type)Protocol: FLIP2 Rhomax by_wild_type
Dataset subset: FLIP2 Rhomax by_wild_type test subset
-0.1463506755340845 spearman
dimensionless · higher

Uncertainty: Not reported

Coverage: 184/184

Rewire evaluation · source checked
Methods, coverage and source

ESM-2 8M frozen residue-mean embeddings + fixed ridge (Rewire) on FLIP2 Rhomax by_wild_type

FLIP2 Rhomax by_wild_type

Aggregation: Not reported

ESM-2 8M frozen residue-mean embeddings + fixed ridge (Rewire): report; ESM-2 8M frozen residue-mean embeddings + fixed ridge (Rewire): audit · metrics.spearman

Source checking is not independent reproduction. Release 2026-09-22-f58a0f1d267f.

Use this model

How it works, versions and access

Related profile: ESM-2. This page retains the exact record and its evaluation context.

This configuration

Frozen features with a fixed alpha-10 ridge head fitted only on training rows; all 184 Rhomax by_wild_type held-out proteins scored. This selected split is not a whole FLIP2 score.

record
ESM-2 8M frozen residue-mean embeddings + fixed ridge (Rewire) (FLIP2 Rhomax by_wild_type)
configuration
Not reported
entity type
Configuration

How it works

How it works

ESM-2 tokenizes an amino-acid sequence and uses a transformer encoder trained to recover masked residues. Self-attention lets each residue representation depend on its sequence context. The released model returns token probabilities and embeddings; a specified pooling rule, task head or complete folding pipeline is needed for a particular biological prediction.

Sources (3)facebookresearch/esm: README.md; facebook/esm2_t33_650M_UR50D: README.md; facebook/esm2_t33_650M_UR50D: config.json · ESM README: Pre-trained Models and Main models; official facebook/esm2_t33_650M_UR50D README licence metadata and config.json
Versions and reproducibility

ESM-2 checkpoint identifiers encode layer count, parameter scale and training-data tag. The checked esm2_t33_650M_UR50D configuration lists max_position_embeddings=1,026. This configuration field includes model positions and is not a claim of training or validated inference on 1,026 amino acids.

Sources (3)facebookresearch/esm: README.md; facebook/esm2_t33_650M_UR50D: README.md; facebook/esm2_t33_650M_UR50D: config.json · ESM README: Pre-trained Models and Main models; official facebook/esm2_t33_650M_UR50D README licence metadata and config.json
Strengths, limitations and unresolved questions

Strengths and limitations

Profile review details

Inspected pinned official documentation, relevant implementation files and named primary-paper sections. Claims are limited to those artifacts. Remaining field extraction and identity conflicts are explicit; no new performance claims, model runs or human review are implied.

Stable record: discovery-model-esm-2

Specifications

Inputs, training, access and other details

Explanatory profile: source reviewed · Automated source review, 2026-09-16. Review applies to the cited claims; unresolved fields are listed below. Numerical results retain their own review status.

Inputs, outputs and configuration
PropertyDescription and evidence
Model typeMasked-token protein transformer encoder
Sources (3)facebookresearch/esm: README.md; facebook/esm2_t33_650M_UR50D: README.md; facebook/esm2_t33_650M_UR50D: config.json · ESM README: Pre-trained Models and Main models; official facebook/esm2_t33_650M_UR50D README licence metadata and config.json
ArchitectureMasked-token protein transformer encoder; the checked 650M checkpoint has 33 layers, hidden width 1,280, 20 attention heads and rotary positional encoding.
Sources (3)facebookresearch/esm: README.md; facebook/esm2_t33_650M_UR50D: README.md; facebook/esm2_t33_650M_UR50D: config.json · ESM README: Pre-trained Models and Main models; official facebook/esm2_t33_650M_UR50D README licence metadata and config.json
InputsSingle amino-acid sequences.
Sources (3)facebookresearch/esm: README.md; facebook/esm2_t33_650M_UR50D: README.md; facebook/esm2_t33_650M_UR50D: config.json · ESM README: Pre-trained Models and Main models; official facebook/esm2_t33_650M_UR50D README licence metadata and config.json
OutputsResidue embeddings, sequence representations and masked-token predictions.
Sources (3)facebookresearch/esm: README.md; facebook/esm2_t33_650M_UR50D: README.md; facebook/esm2_t33_650M_UR50D: config.json · ESM README: Pre-trained Models and Main models; official facebook/esm2_t33_650M_UR50D README licence metadata and config.json
ParametersReleased scales: 8M, 35M, 150M, 650M, 3B and 15B.
Sources (3)facebookresearch/esm: README.md; facebook/esm2_t33_650M_UR50D: README.md; facebook/esm2_t33_650M_UR50D: config.json · ESM README: Pre-trained Models and Main models; official facebook/esm2_t33_650M_UR50D README licence metadata and config.json
Known versionsESM-2 checkpoint identifiers encode layer count, parameter scale and training-data tag.
Sources (3)facebookresearch/esm: README.md; facebook/esm2_t33_650M_UR50D: README.md; facebook/esm2_t33_650M_UR50D: config.json · ESM README: Pre-trained Models and Main models; official facebook/esm2_t33_650M_UR50D README licence metadata and config.json
Training dataUniRef50 clusters with UniRef90 sampling; the pretrained-model table labels UR50/D 2021_04.
Sources (3)facebookresearch/esm: README.md; facebook/esm2_t33_650M_UR50D: README.md; facebook/esm2_t33_650M_UR50D: config.json · ESM README: Pre-trained Models and Main models; official facebook/esm2_t33_650M_UR50D README licence metadata and config.json
Training cutoffThe pretrained-model table identifies training-data release UR50/D 2021_04; a corpus release date is not necessarily a last-deposited-sequence cutoff.
Sources (3)facebookresearch/esm: README.md; facebook/esm2_t33_650M_UR50D: README.md; facebook/esm2_t33_650M_UR50D: config.json · ESM README: Pre-trained Models and Main models; official facebook/esm2_t33_650M_UR50D README licence metadata and config.json
Context limitsThe checked esm2_t33_650M_UR50D configuration lists max_position_embeddings=1,026. This configuration field includes model positions and is not a claim of training or validated inference on 1,026 amino acids.
Sources (3)facebookresearch/esm: README.md; facebook/esm2_t33_650M_UR50D: README.md; facebook/esm2_t33_650M_UR50D: config.json · ESM README: Pre-trained Models and Main models; official facebook/esm2_t33_650M_UR50D README licence metadata and config.json
Weights licenceThe official facebook/esm2_t33_650M_UR50D model card declares MIT; this is the inspected checkpoint, not a licence inference from source code.
Sources (3)facebookresearch/esm: README.md; facebook/esm2_t33_650M_UR50D: README.md; facebook/esm2_t33_650M_UR50D: config.json · ESM README: Pre-trained Models and Main models; official facebook/esm2_t33_650M_UR50D README licence metadata and config.json
AccessOfficial project documentation and implementation: https://github.com/facebookresearch/esm
Sources (3)facebookresearch/esm: README.md; facebook/esm2_t33_650M_UR50D: README.md; facebook/esm2_t33_650M_UR50D: config.json · ESM README: Pre-trained Models and Main models; official facebook/esm2_t33_650M_UR50D README licence metadata and config.json
Code licenceMIT
Sourcesfacebookresearch/esm: LICENSE · LICENSE: licence text

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 2026-09-22-f58a0f1d267f
Property and statementOriginal source and locationReview and provenance
Relationship: family

discovery-model-esm-2

Individual claims
ESM-2 8M frozen residue-mean embeddings + fixed ridge (Rewire): report

Original source ↗

model_configuration.checkpoint; model.name

Version: 1663d1f04b2bbd6dfcff77fea78129d30b0de191
Retrieved: 2026-09-22T21:18:22Z

source checked

automated execution evidence review · 2026-09-22T21:18:22Z

Audit details

Field: links:family:discovery-model-esm-2

Claim: rewire-local-20260921-configuration-esm2-8m-family

Source artifact SHA-256: d6df9e916e77872112571b404265e918276974c79dcbe25721b3eca0eae4e51a

Hash scope: Hash scope not separately documented; inspect source record

Inspected artifact

Sources and history

View linked audit checks and correction history

Release 2026-09-22-f58a0f1d267f · Record review: source checked

2 source records and release historyDownload this release
Technical metadata and extraction receipts

Stable ID: rewire-local-20260921-configuration-esm2-8m

areas
proteins-complexes
source locator
model; model_configuration; protocol_configuration.embedding_probe; execution
configuration
checkpoint: esm2_t6_8M_UR50D; device: cpu; dimension: 320; encoder frozen: true; head: alpha: 10; feature scaling: none; fit intercept: true; max iter: 1000000; name: rewire-frozen-embedding-ridge-v1; refit: train only; relationship to published baseline: Frozen-embedding extension of upstream alpha-10 ridge. The published baseline uses one-hot features and scales targets using train plus validation rows. This extension is not that published baseline.; solver: auto; target scaling: StandardScaler fitted only on train labels; tol: 0.00001; validation use: none; fixed hyperparameters; pooling: mean over residues excluding BOS/EOS and padding; representation layer: 6; scope: one complete Rhomax split, not FLIP2 suite; torch seed: 0; torch threads: 4; validation used: false
execution
adapter: ESM2Embeddings; adapter provenance: checkpoint sha256: 46f002a9870c9bdecd0ea887acb1f9a38a6b561e8f8bf8a6990b679b9d31b928; checkpoint url: https://dl.fbaipublicfiles.com/fair-esm/models/esm2_t6_8M_UR50D.pt; device: cpu; fair esm version: 2.0.0; implementation sha256: 1f475a084964698145fb07f7d0cd851a486d06a508e266904160d6d6ec9ef9bb; maximum sequence length: 1022; model: esm2_t6_8M_UR50D; published result reproduction: false; reference: https://github.com/facebookresearch/esm#usage; representation layer: 6; strategy: frozen_final_layer_residue_mean_excluding_special_tokens; training overlap: unreported; UniRef50 pretraining may overlap benchmark proteins; batch size: 4; fitting: protocol_owned_probe; inference and fit seconds: 17.59812141599832; mode: local_adapter; prediction type: embedding
model
name: ESM-2 8M frozen residue-mean embeddings + fixed ridge (Rewire); training overlap: UniRef50 pretraining overlap is unreported; the linear head fits only archived training rows
environment
architecture: arm64; container digest: unreported; container runtime: unreported; platform: macOS-26.6.2-arm64-arm-64bit; python: 3.11.13; rewirebench: 0.4.0; sdk code sha256: 9350e367b20ec48d9ce1195cb5b73c4bd98ce8530a2d1f7ba2157fc8be75fa34; sif sha256: unreported
training overlap
UniRef50 pretraining overlap is unreported; the linear head fits only archived training rows
method role
frozen pretrained encoder plus trained regression head
Related records

Suggest a correction