Platonic Protein Space
Data for testing whether protein, description and annotation embeddings share one geometry.
Viewer • Updated • 67.4k • 36Note Translator training and alignment data
GleghornLab/vec2vec_embeddings
Updated • 72Note Frozen encoder embeddings of PLAT
Synthyra/vector_embeddings
Updated • 31Note ESM2 and control embeddings for the downstream probes
GleghornLab/BP_reg
Viewer • Updated • 32.5k • 13Note Downstream task: GO biological process
GleghornLab/CC_reg
Viewer • Updated • 32.5k • 8Note Downstream task: GO cellular component
GleghornLab/MF_reg
Viewer • Updated • 32.5k • 11Note Downstream task: GO molecular function
GleghornLab/EC_reg
Viewer • Updated • 16.2k • 12Note Downstream task: EC number
GleghornLab/SL_13
Viewer • Updated • 34.6k • 22Note Downstream task: subcellular localization
GleghornLab/solubility_prediction
Viewer • Updated • 69.4k • 43Note Downstream task: solubility
GleghornLab/temperature_stability
Viewer • Updated • 419k • 27Note Downstream task: temperature stability
Synthyra/bernett_gold_ppi
Viewer • Updated • 275k • 95 • 2Note Downstream task: protein-protein interaction
Synthyra/ESM2-8M
Fill-Mask • 7.52M • Updated • 715 • 4Note Encoder: ESM2 8M
Synthyra/ESM2-35M
Fill-Mask • 33.5M • Updated • 4.69k • 2Note Encoder: ESM2 35M
Synthyra/ESM2-150M
Fill-Mask • 0.1B • Updated • 328 • 1Note Encoder: ESM2 150M
Synthyra/ESM2-650M
Fill-Mask • 0.7B • Updated • 1.27k • 1Note Encoder: ESM2 650M
Synthyra/ESM2-3B
Fill-Mask • 3B • Updated • 171 • 1Note Encoder: ESM2 3B
answerdotai/ModernBERT-base
Fill-Mask • 0.1B • Updated • 3.3M • 1.15kNote Encoder: sequence and description
answerdotai/ModernBERT-large
Fill-Mask • 0.4B • Updated • 680k • 500Note Encoder: sequence and description
openai/gpt-oss-20b
Text Generation • 21B • Updated • 6.13M • • 5.26kNote Encoder: sequence and description
openai/gpt-oss-120b
Text Generation • 117B • Updated • 2.63M • • 5.47kNote Encoder: sequence and description