Evotuned ESM-2 650M, forced substitution
The same family-adapted ESM-2 650M infilling run with the template residue masked out of the output logits at every selected position, so every selected position changes in a single pass.
Overview
An explanatory profile has not yet been reviewed for this record.
Consult the linked sources for architecture or protocol details. Missing evidence is not evidence of a missing capability.
Model Capabilities & Hardware Requirements
Deployment facts recorded for Evotuned ESM-2 650M, forced substitution. Fields the record does not carry are marked not recorded.
Evaluated Checkpoint: As reimplemented and scored in version 1 of the paper
Evaluations and results
20 evaluations · 20 results. Different protocols are not a single leaderboard.
Sorted by Covariance agreement (residue covariance agreement against a 300-sequence natural homolog ceiling) (higher is better). The best value in each column is highlighted. Decimals are rounded for display; each value links to the printed value and its source.
| Tested configuration | Protocol and dataset | Covariance agreement (residue covariance agreement against a 300-sequence natural homolog ceiling) | Jensen shannon divergence (per-position formulation chosen by the EditJumps authors, against a 200-sequence natural holdout) | Kullback leibler divergence (per-position amino-acid composition against a 200-sequence natural holdout) | Kullback leibler divergence (pooled amino-acid composition against a 200-sequence natural holdout) | Log likelihood (profile log-likelihood of the generated sequences under the family alignment profile) | Maximum mean discrepancy (k-mer spectrum MMD between the generated set and a 200-sequence natural holdout, unnormalised) | Mutual information product agreement (mutual information product agreement against a 300-sequence natural homolog ceiling) | Edit count (mean realised edits per generated sequence, 400 generated sequences per family) | Entropy delta (positional entropy difference from the natural holdout; the target value is zero) | Levenshtein distance (mean pairwise Levenshtein distance within the generated set, pooled over templates) |
|---|---|---|---|---|---|---|---|---|---|---|---|
| Evotuned ESM-2 650M, forced substitutionEvaluation | EditJumps section 4.2 frame, 200 references: cov. agr., HER2-VH reference holdout, 200 homologs | 0.91 (best in column) | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported |
| Evotuned ESM-2 650M, forced substitutionEvaluation | EditJumps section 4.2 frame, 200 references: cov. agr., Ty1 reference holdout, 200 homologs | 0.87 | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported |
| Evotuned ESM-2 650M, forced substitutionEvaluation | EditJumps section 4.2 frame, 200 references: edits, HER2-VH reference holdout, 200 homologs | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | 4 | Not reported | Not reported |
| Evotuned ESM-2 650M, forced substitutionEvaluation | EditJumps section 4.2 frame, 200 references: edits, Ty1 reference holdout, 200 homologs | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | 5 | Not reported | Not reported |
| Evotuned ESM-2 650M, forced substitutionEvaluation | EditJumps section 4.2 frame, 200 references: ΔH, HER2-VH reference holdout, 200 homologs | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | 0.1 | Not reported |
| Evotuned ESM-2 650M, forced substitutionEvaluation | EditJumps section 4.2 frame, 200 references: ΔH, Ty1 reference holdout, 200 homologs | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | 0.09 | Not reported |
| Evotuned ESM-2 650M, forced substitutionEvaluation | EditJumps section 4.2 frame, 200 references: JS pos., HER2-VH reference holdout, 200 homologs | Not reported | 1.21 (best in column) | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported |
| Evotuned ESM-2 650M, forced substitutionEvaluation | EditJumps section 4.2 frame, 200 references: JS pos., Ty1 reference holdout, 200 homologs | Not reported | 1.4 | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported |
| Evotuned ESM-2 650M, forced substitutionEvaluation | EditJumps section 4.2 frame, 200 references: KL pool., HER2-VH reference holdout, 200 homologs | Not reported | Not reported | Not reported | 0.59 | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported |
| Evotuned ESM-2 650M, forced substitutionEvaluation | EditJumps section 4.2 frame, 200 references: KL pool., Ty1 reference holdout, 200 homologs | Not reported | Not reported | Not reported | 0.53 (best in column) | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported |
| Evotuned ESM-2 650M, forced substitutionEvaluation | EditJumps section 4.2 frame, 200 references: KL pos., HER2-VH reference holdout, 200 homologs | Not reported | Not reported | 0.07 (best in column) | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported |
| Evotuned ESM-2 650M, forced substitutionEvaluation | EditJumps section 4.2 frame, 200 references: KL pos., Ty1 reference holdout, 200 homologs | Not reported | Not reported | 0.09 | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported |
| Evotuned ESM-2 650M, forced substitutionEvaluation | EditJumps section 4.2 frame, 200 references: MIP agr., HER2-VH reference holdout, 200 homologs | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | 0.81 (best in column) | Not reported | Not reported | Not reported |
| Evotuned ESM-2 650M, forced substitutionEvaluation | EditJumps section 4.2 frame, 200 references: MIP agr., Ty1 reference holdout, 200 homologs | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | 0.74 | Not reported | Not reported | Not reported |
| Evotuned ESM-2 650M, forced substitutionEvaluation | EditJumps section 4.2 frame, 200 references: pairw. Lev., HER2-VH reference holdout, 200 homologs | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | 25 |
| Evotuned ESM-2 650M, forced substitutionEvaluation | EditJumps section 4.2 frame, 200 references: pairw. Lev., Ty1 reference holdout, 200 homologs | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | 25.2 |
| Evotuned ESM-2 650M, forced substitutionEvaluation | EditJumps section 4.2 frame, 200 references: prof. LL, HER2-VH reference holdout, 200 homologs | Not reported | Not reported | Not reported | Not reported | -62.2 (best in column) | Not reported | Not reported | Not reported | Not reported | Not reported |
| Evotuned ESM-2 650M, forced substitutionEvaluation | EditJumps section 4.2 frame, 200 references: prof. LL, Ty1 reference holdout, 200 homologs | Not reported | Not reported | Not reported | Not reported | -67.5 | Not reported | Not reported | Not reported | Not reported | Not reported |
| Evotuned ESM-2 650M, forced substitutionEvaluation | EditJumps section 4.2 frame, 200 references: MMD, HER2-VH reference holdout, 200 homologs | Not reported | Not reported | Not reported | Not reported | Not reported | 1.15 | Not reported | Not reported | Not reported | Not reported |
| Evotuned ESM-2 650M, forced substitutionEvaluation | EditJumps section 4.2 frame, 200 references: MMD, Ty1 reference holdout, 200 homologs | Not reported | Not reported | Not reported | Not reported | Not reported | 1.09 (best in column) | Not reported | Not reported | Not reported | Not reported |
All 20 result rows with coverage, uncertainty and sources
Filter evaluations
Applied filters: All linked evaluations
| Tested configuration | Protocol and dataset | Finding | Evidence and details |
|---|---|---|---|
| Configuration: Evotuned ESM-2 650M, forced substitution | Protocol: EditJumps section 4.2 frame, 200 references: cov. agr. Dataset subset: HER2-VH reference holdout, 200 homologs | 0.91 Covariance agreement (residue covariance agreement against a 300-sequence natural homolog ceiling) unitless · higher Uncertainty: Not reported by the source: The table prints a single value with no interval. Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceEvotune-f on HER2-VH, cov. agr., 200 references One frame for every arm and both families: 20 templates per family and 20 variants per template, so 400 generated sequences, scored against a 200-sequence held-out natural homolog set, with the agreement ceiling computed on 300 real homologs. Sampling runs Euler tau-leaping over 50 steps. Every arm is held to a matched realised edit budget of about 4 to 5 edits per sequence. Aggregation: Not reported When Edit Flows are Edit Jumps: replicating Edit Flows and EvoFlows (arXiv v1) · Table 2, family block "her2vh", row "Evotune-f", column "cov. agr." |
| Configuration: Evotuned ESM-2 650M, forced substitution | Protocol: EditJumps section 4.2 frame, 200 references: cov. agr. Dataset subset: Ty1 reference holdout, 200 homologs | 0.87 Covariance agreement (residue covariance agreement against a 300-sequence natural homolog ceiling) unitless · higher Uncertainty: Not reported by the source: The table prints a single value with no interval. Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceEvotune-f on Ty1, cov. agr., 200 references One frame for every arm and both families: 20 templates per family and 20 variants per template, so 400 generated sequences, scored against a 200-sequence held-out natural homolog set, with the agreement ceiling computed on 300 real homologs. Sampling runs Euler tau-leaping over 50 steps. Every arm is held to a matched realised edit budget of about 4 to 5 edits per sequence. Aggregation: Not reported When Edit Flows are Edit Jumps: replicating Edit Flows and EvoFlows (arXiv v1) · Table 2, family block "ty1", row "Evotune-f", column "cov. agr." |
| Configuration: Evotuned ESM-2 650M, forced substitution | Protocol: EditJumps section 4.2 frame, 200 references: edits Dataset subset: HER2-VH reference holdout, 200 homologs | 4 Edit count (mean realised edits per generated sequence, 400 generated sequences per family) count · unknown Uncertainty: Not reported by the source: The table prints a single value with no interval. Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceEvotune-f on HER2-VH, edits, 200 references One frame for every arm and both families: 20 templates per family and 20 variants per template, so 400 generated sequences, scored against a 200-sequence held-out natural homolog set, with the agreement ceiling computed on 300 real homologs. Sampling runs Euler tau-leaping over 50 steps. Every arm is held to a matched realised edit budget of about 4 to 5 edits per sequence. Aggregation: Not reported When Edit Flows are Edit Jumps: replicating Edit Flows and EvoFlows (arXiv v1) · Table 2, family block "her2vh", row "Evotune-f", column "edits" |
| Configuration: Evotuned ESM-2 650M, forced substitution | Protocol: EditJumps section 4.2 frame, 200 references: edits Dataset subset: Ty1 reference holdout, 200 homologs | 5 Edit count (mean realised edits per generated sequence, 400 generated sequences per family) count · unknown Uncertainty: Not reported by the source: The table prints a single value with no interval. Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceEvotune-f on Ty1, edits, 200 references One frame for every arm and both families: 20 templates per family and 20 variants per template, so 400 generated sequences, scored against a 200-sequence held-out natural homolog set, with the agreement ceiling computed on 300 real homologs. Sampling runs Euler tau-leaping over 50 steps. Every arm is held to a matched realised edit budget of about 4 to 5 edits per sequence. Aggregation: Not reported When Edit Flows are Edit Jumps: replicating Edit Flows and EvoFlows (arXiv v1) · Table 2, family block "ty1", row "Evotune-f", column "edits" |
| Configuration: Evotuned ESM-2 650M, forced substitution | Protocol: EditJumps section 4.2 frame, 200 references: ΔH Dataset subset: HER2-VH reference holdout, 200 homologs | 0.1 Entropy delta (positional entropy difference from the natural holdout; the target value is zero) unitless · unknown Uncertainty: Not reported by the source: The table prints a single value with no interval. Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceEvotune-f on HER2-VH, ΔH, 200 references One frame for every arm and both families: 20 templates per family and 20 variants per template, so 400 generated sequences, scored against a 200-sequence held-out natural homolog set, with the agreement ceiling computed on 300 real homologs. Sampling runs Euler tau-leaping over 50 steps. Every arm is held to a matched realised edit budget of about 4 to 5 edits per sequence. Aggregation: Not reported When Edit Flows are Edit Jumps: replicating Edit Flows and EvoFlows (arXiv v1) · Table 2, family block "her2vh", row "Evotune-f", column "ΔH" |
| Configuration: Evotuned ESM-2 650M, forced substitution | Protocol: EditJumps section 4.2 frame, 200 references: ΔH Dataset subset: Ty1 reference holdout, 200 homologs | 0.09 Entropy delta (positional entropy difference from the natural holdout; the target value is zero) unitless · unknown Uncertainty: Not reported by the source: The table prints a single value with no interval. Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceEvotune-f on Ty1, ΔH, 200 references One frame for every arm and both families: 20 templates per family and 20 variants per template, so 400 generated sequences, scored against a 200-sequence held-out natural homolog set, with the agreement ceiling computed on 300 real homologs. Sampling runs Euler tau-leaping over 50 steps. Every arm is held to a matched realised edit budget of about 4 to 5 edits per sequence. Aggregation: Not reported When Edit Flows are Edit Jumps: replicating Edit Flows and EvoFlows (arXiv v1) · Table 2, family block "ty1", row "Evotune-f", column "ΔH" |
| Configuration: Evotuned ESM-2 650M, forced substitution | Protocol: EditJumps section 4.2 frame, 200 references: JS pos. Dataset subset: HER2-VH reference holdout, 200 homologs | 1.21 Jensen shannon divergence (per-position formulation chosen by the EditJumps authors, against a 200-sequence natural holdout) unitless · lower Uncertainty: Not reported by the source: The table prints a single value with no interval. Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceEvotune-f on HER2-VH, JS pos., 200 references One frame for every arm and both families: 20 templates per family and 20 variants per template, so 400 generated sequences, scored against a 200-sequence held-out natural homolog set, with the agreement ceiling computed on 300 real homologs. Sampling runs Euler tau-leaping over 50 steps. Every arm is held to a matched realised edit budget of about 4 to 5 edits per sequence. Aggregation: Not reported When Edit Flows are Edit Jumps: replicating Edit Flows and EvoFlows (arXiv v1) · Table 2, family block "her2vh", row "Evotune-f", column "JS pos." |
| Configuration: Evotuned ESM-2 650M, forced substitution | Protocol: EditJumps section 4.2 frame, 200 references: JS pos. Dataset subset: Ty1 reference holdout, 200 homologs | 1.4 Jensen shannon divergence (per-position formulation chosen by the EditJumps authors, against a 200-sequence natural holdout) unitless · lower Uncertainty: Not reported by the source: The table prints a single value with no interval. Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceEvotune-f on Ty1, JS pos., 200 references One frame for every arm and both families: 20 templates per family and 20 variants per template, so 400 generated sequences, scored against a 200-sequence held-out natural homolog set, with the agreement ceiling computed on 300 real homologs. Sampling runs Euler tau-leaping over 50 steps. Every arm is held to a matched realised edit budget of about 4 to 5 edits per sequence. Aggregation: Not reported When Edit Flows are Edit Jumps: replicating Edit Flows and EvoFlows (arXiv v1) · Table 2, family block "ty1", row "Evotune-f", column "JS pos." |
| Configuration: Evotuned ESM-2 650M, forced substitution | Protocol: EditJumps section 4.2 frame, 200 references: KL pool. Dataset subset: HER2-VH reference holdout, 200 homologs | 0.59 Kullback leibler divergence (pooled amino-acid composition against a 200-sequence natural holdout) unitless · lower Uncertainty: Not reported by the source: The table prints a single value with no interval. Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceEvotune-f on HER2-VH, KL pool., 200 references One frame for every arm and both families: 20 templates per family and 20 variants per template, so 400 generated sequences, scored against a 200-sequence held-out natural homolog set, with the agreement ceiling computed on 300 real homologs. Sampling runs Euler tau-leaping over 50 steps. Every arm is held to a matched realised edit budget of about 4 to 5 edits per sequence. Aggregation: Not reported When Edit Flows are Edit Jumps: replicating Edit Flows and EvoFlows (arXiv v1) · Table 2, family block "her2vh", row "Evotune-f", column "KL pool." |
| Configuration: Evotuned ESM-2 650M, forced substitution | Protocol: EditJumps section 4.2 frame, 200 references: KL pool. Dataset subset: Ty1 reference holdout, 200 homologs | 0.53 Kullback leibler divergence (pooled amino-acid composition against a 200-sequence natural holdout) unitless · lower Uncertainty: Not reported by the source: The table prints a single value with no interval. Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceEvotune-f on Ty1, KL pool., 200 references One frame for every arm and both families: 20 templates per family and 20 variants per template, so 400 generated sequences, scored against a 200-sequence held-out natural homolog set, with the agreement ceiling computed on 300 real homologs. Sampling runs Euler tau-leaping over 50 steps. Every arm is held to a matched realised edit budget of about 4 to 5 edits per sequence. Aggregation: Not reported When Edit Flows are Edit Jumps: replicating Edit Flows and EvoFlows (arXiv v1) · Table 2, family block "ty1", row "Evotune-f", column "KL pool." |
| Configuration: Evotuned ESM-2 650M, forced substitution | Protocol: EditJumps section 4.2 frame, 200 references: KL pos. Dataset subset: HER2-VH reference holdout, 200 homologs | 0.07 Kullback leibler divergence (per-position amino-acid composition against a 200-sequence natural holdout) unitless · lower Uncertainty: Not reported by the source: The table prints a single value with no interval. Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceEvotune-f on HER2-VH, KL pos., 200 references One frame for every arm and both families: 20 templates per family and 20 variants per template, so 400 generated sequences, scored against a 200-sequence held-out natural homolog set, with the agreement ceiling computed on 300 real homologs. Sampling runs Euler tau-leaping over 50 steps. Every arm is held to a matched realised edit budget of about 4 to 5 edits per sequence. Aggregation: Not reported When Edit Flows are Edit Jumps: replicating Edit Flows and EvoFlows (arXiv v1) · Table 2, family block "her2vh", row "Evotune-f", column "KL pos." |
| Configuration: Evotuned ESM-2 650M, forced substitution | Protocol: EditJumps section 4.2 frame, 200 references: KL pos. Dataset subset: Ty1 reference holdout, 200 homologs | 0.09 Kullback leibler divergence (per-position amino-acid composition against a 200-sequence natural holdout) unitless · lower Uncertainty: Not reported by the source: The table prints a single value with no interval. Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceEvotune-f on Ty1, KL pos., 200 references One frame for every arm and both families: 20 templates per family and 20 variants per template, so 400 generated sequences, scored against a 200-sequence held-out natural homolog set, with the agreement ceiling computed on 300 real homologs. Sampling runs Euler tau-leaping over 50 steps. Every arm is held to a matched realised edit budget of about 4 to 5 edits per sequence. Aggregation: Not reported When Edit Flows are Edit Jumps: replicating Edit Flows and EvoFlows (arXiv v1) · Table 2, family block "ty1", row "Evotune-f", column "KL pos." |
| Configuration: Evotuned ESM-2 650M, forced substitution | Protocol: EditJumps section 4.2 frame, 200 references: MIP agr. Dataset subset: HER2-VH reference holdout, 200 homologs | 0.81 Mutual information product agreement (mutual information product agreement against a 300-sequence natural homolog ceiling) unitless · higher Uncertainty: Not reported by the source: The table prints a single value with no interval. Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceEvotune-f on HER2-VH, MIP agr., 200 references One frame for every arm and both families: 20 templates per family and 20 variants per template, so 400 generated sequences, scored against a 200-sequence held-out natural homolog set, with the agreement ceiling computed on 300 real homologs. Sampling runs Euler tau-leaping over 50 steps. Every arm is held to a matched realised edit budget of about 4 to 5 edits per sequence. Aggregation: Not reported When Edit Flows are Edit Jumps: replicating Edit Flows and EvoFlows (arXiv v1) · Table 2, family block "her2vh", row "Evotune-f", column "MIP agr." |
| Configuration: Evotuned ESM-2 650M, forced substitution | Protocol: EditJumps section 4.2 frame, 200 references: MIP agr. Dataset subset: Ty1 reference holdout, 200 homologs | 0.74 Mutual information product agreement (mutual information product agreement against a 300-sequence natural homolog ceiling) unitless · higher Uncertainty: Not reported by the source: The table prints a single value with no interval. Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceEvotune-f on Ty1, MIP agr., 200 references One frame for every arm and both families: 20 templates per family and 20 variants per template, so 400 generated sequences, scored against a 200-sequence held-out natural homolog set, with the agreement ceiling computed on 300 real homologs. Sampling runs Euler tau-leaping over 50 steps. Every arm is held to a matched realised edit budget of about 4 to 5 edits per sequence. Aggregation: Not reported When Edit Flows are Edit Jumps: replicating Edit Flows and EvoFlows (arXiv v1) · Table 2, family block "ty1", row "Evotune-f", column "MIP agr." |
| Configuration: Evotuned ESM-2 650M, forced substitution | Protocol: EditJumps section 4.2 frame, 200 references: pairw. Lev. Dataset subset: HER2-VH reference holdout, 200 homologs | 25 Levenshtein distance (mean pairwise Levenshtein distance within the generated set, pooled over templates) count · unknown Uncertainty: Not reported by the source: The table prints a single value with no interval. Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceEvotune-f on HER2-VH, pairw. Lev., 200 references One frame for every arm and both families: 20 templates per family and 20 variants per template, so 400 generated sequences, scored against a 200-sequence held-out natural homolog set, with the agreement ceiling computed on 300 real homologs. Sampling runs Euler tau-leaping over 50 steps. Every arm is held to a matched realised edit budget of about 4 to 5 edits per sequence. Aggregation: Not reported When Edit Flows are Edit Jumps: replicating Edit Flows and EvoFlows (arXiv v1) · Table 2, family block "her2vh", row "Evotune-f", column "pairw. Lev." |
| Configuration: Evotuned ESM-2 650M, forced substitution | Protocol: EditJumps section 4.2 frame, 200 references: pairw. Lev. Dataset subset: Ty1 reference holdout, 200 homologs | 25.2 Levenshtein distance (mean pairwise Levenshtein distance within the generated set, pooled over templates) count · unknown Uncertainty: Not reported by the source: The table prints a single value with no interval. Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceEvotune-f on Ty1, pairw. Lev., 200 references One frame for every arm and both families: 20 templates per family and 20 variants per template, so 400 generated sequences, scored against a 200-sequence held-out natural homolog set, with the agreement ceiling computed on 300 real homologs. Sampling runs Euler tau-leaping over 50 steps. Every arm is held to a matched realised edit budget of about 4 to 5 edits per sequence. Aggregation: Not reported When Edit Flows are Edit Jumps: replicating Edit Flows and EvoFlows (arXiv v1) · Table 2, family block "ty1", row "Evotune-f", column "pairw. Lev." |
| Configuration: Evotuned ESM-2 650M, forced substitution | Protocol: EditJumps section 4.2 frame, 200 references: prof. LL Dataset subset: HER2-VH reference holdout, 200 homologs | -62.2 Log likelihood (profile log-likelihood of the generated sequences under the family alignment profile) unitless · higher Uncertainty: Not reported by the source: The table prints a single value with no interval. Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceEvotune-f on HER2-VH, prof. LL, 200 references One frame for every arm and both families: 20 templates per family and 20 variants per template, so 400 generated sequences, scored against a 200-sequence held-out natural homolog set, with the agreement ceiling computed on 300 real homologs. Sampling runs Euler tau-leaping over 50 steps. Every arm is held to a matched realised edit budget of about 4 to 5 edits per sequence. Aggregation: Not reported When Edit Flows are Edit Jumps: replicating Edit Flows and EvoFlows (arXiv v1) · Table 2, family block "her2vh", row "Evotune-f", column "prof. LL" |
| Configuration: Evotuned ESM-2 650M, forced substitution | Protocol: EditJumps section 4.2 frame, 200 references: prof. LL Dataset subset: Ty1 reference holdout, 200 homologs | -67.5 Log likelihood (profile log-likelihood of the generated sequences under the family alignment profile) unitless · higher Uncertainty: Not reported by the source: The table prints a single value with no interval. Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceEvotune-f on Ty1, prof. LL, 200 references One frame for every arm and both families: 20 templates per family and 20 variants per template, so 400 generated sequences, scored against a 200-sequence held-out natural homolog set, with the agreement ceiling computed on 300 real homologs. Sampling runs Euler tau-leaping over 50 steps. Every arm is held to a matched realised edit budget of about 4 to 5 edits per sequence. Aggregation: Not reported When Edit Flows are Edit Jumps: replicating Edit Flows and EvoFlows (arXiv v1) · Table 2, family block "ty1", row "Evotune-f", column "prof. LL" |
| Configuration: Evotuned ESM-2 650M, forced substitution | Protocol: EditJumps section 4.2 frame, 200 references: MMD Dataset subset: HER2-VH reference holdout, 200 homologs | 1.15 Maximum mean discrepancy (k-mer spectrum MMD between the generated set and a 200-sequence natural holdout, unnormalised) unitless · lower Uncertainty: Not reported by the source: The table prints a single value with no interval. Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceEvotune-f on HER2-VH, MMD, 200 references One frame for every arm and both families: 20 templates per family and 20 variants per template, so 400 generated sequences, scored against a 200-sequence held-out natural homolog set, with the agreement ceiling computed on 300 real homologs. Sampling runs Euler tau-leaping over 50 steps. Every arm is held to a matched realised edit budget of about 4 to 5 edits per sequence. Aggregation: Not reported When Edit Flows are Edit Jumps: replicating Edit Flows and EvoFlows (arXiv v1) · Table 2, family block "her2vh", row "Evotune-f", column "MMD" |
| Configuration: Evotuned ESM-2 650M, forced substitution | Protocol: EditJumps section 4.2 frame, 200 references: MMD Dataset subset: Ty1 reference holdout, 200 homologs | 1.09 Maximum mean discrepancy (k-mer spectrum MMD between the generated set and a 200-sequence natural holdout, unnormalised) unitless · lower Uncertainty: Not reported by the source: The table prints a single value with no interval. Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceEvotune-f on Ty1, MMD, 200 references One frame for every arm and both families: 20 templates per family and 20 variants per template, so 400 generated sequences, scored against a 200-sequence held-out natural homolog set, with the agreement ceiling computed on 300 real homologs. Sampling runs Euler tau-leaping over 50 steps. Every arm is held to a matched realised edit budget of about 4 to 5 edits per sequence. Aggregation: Not reported When Edit Flows are Edit Jumps: replicating Edit Flows and EvoFlows (arXiv v1) · Table 2, family block "ty1", row "Evotune-f", column "MMD" |
Source checking is not independent reproduction. Release 2026-11-21-c8bea25e5abb.
Use this configuration
How it works, versions and access
Additional methodological details, versions and access terms have not yet been extracted into a structured profile for this configuration. Consult the linked sources and evaluation records below.
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.
0 evidence rows matching the loaded filters
| Property and statement | Original source and location | Review and provenance |
|---|
No evidence rows match these filters. Choose another scope or clear the search.
Sources and history
Release 2026-11-21-c8bea25e5abb · Record review: source checked
1 source record and release history
- When Edit Flows are Edit Jumps: replicating Edit Flows and EvoFlows (arXiv v1) · Original source · arXiv:2609.18745v1, 16 September 2026
Technical metadata and record identifiers
Technical metadata for this record is tracked under stable ID editjumps-20261115-config-evotuning-650m-forced.
Source locator: Section 3 'The evotuning baseline'; Appendix A item 4
Related records
- configuration of: Evotuning
- uses model: ESM-2
- system: Evotune-f on HER2-VH, cov. agr., 200 references
- system: Evotune-f on Ty1, cov. agr., 200 references
- system: Evotune-f on HER2-VH, edits, 200 references
- system: Evotune-f on Ty1, edits, 200 references
- system: Evotune-f on HER2-VH, ΔH, 200 references
- system: Evotune-f on Ty1, ΔH, 200 references
- system: Evotune-f on HER2-VH, JS pos., 200 references
- system: Evotune-f on Ty1, JS pos., 200 references
- system: Evotune-f on HER2-VH, KL pool., 200 references
- system: Evotune-f on Ty1, KL pool., 200 references
- system: Evotune-f on HER2-VH, KL pos., 200 references
- system: Evotune-f on Ty1, KL pos., 200 references
- system: Evotune-f on HER2-VH, MIP agr., 200 references
- system: Evotune-f on Ty1, MIP agr., 200 references
- system: Evotune-f on HER2-VH, pairw. Lev., 200 references
- system: Evotune-f on Ty1, pairw. Lev., 200 references
- system: Evotune-f on HER2-VH, prof. LL, 200 references
- system: Evotune-f on Ty1, prof. LL, 200 references
- system: Evotune-f on HER2-VH, MMD, 200 references
- system: Evotune-f on Ty1, MMD, 200 references
How to cite
Evotuned ESM-2 650M, forced substitution. Rewire biological model benchmark database, release 2026-11-21-c8bea25e5abb. https://benchmarks.rewirebio.io/database/configuration/editjumps-20261115-config-evotuning-650m-forced/
Release 2026-11-21-c8bea25e5abb. Rankings and values change between releases, so keep the release id in the citation. Cite the original publication as well for any value you reuse: each result links its source.
The compilation and annotations made by Rewire in this release are offered under CC BY 4.0; values and text taken from cited sources stay under the terms of those sources and are not covered by this licence. Data licence