A retraining of a TokenGT, embedding space GFlowNets GoT reasoning 4x4M-Soft-MoE model with tropical ring attention and geometric algebra constraints
Amelie Schreiber
AmelieSchreiber
AI & ML interests
Diffusion and flow matching models for proteins, small molecules, DNA, and RNA, protein language models, machine translation, equivariant attention mechanisms, LoRA, and QLoRA.
Recent Activity
commentedon a paper about 15 hours ago
Dynamic Latent Routing updated a collection about 21 hours ago
ToricGTOrganizations
QLoRAs for various ESM-2 models for predicting binding sites of protein sequences.
ESM-2 models finetuned for generating peptide binders and predicting protein-protein interactions using MLM capabilities. None of the models work well
ESMBind (or ESMB) are LoRA finetuned ESM-2 models for predicting general binding sites and active sites of proteins based on sequence alone.
-
AmelieSchreiber/esm2_t12_35M_lora_binding_sites_770K_v1
Updated • 6 -
AmelieSchreiber/esm2_t12_35M_lora_binding_sites_v2_cp3
Token Classification • Updated • 17 • 6 -
AmelieSchreiber/esm2_t12_35M_lora_binding_sites_v2_cp2
Token Classification • Updated • 10 -
AmelieSchreiber/esm2_t12_35M_lora_binding_sites_v2_cp1
Token Classification • Updated • 8
This is a collection of LoRA and QLoRA finetuned ESM-2 models and datasets for predicting post translational modification sites on proteins.
A retraining of a TokenGT, embedding space GFlowNets GoT reasoning 4x4M-Soft-MoE model with tropical ring attention and geometric algebra constraints
ESMBind (or ESMB) are LoRA finetuned ESM-2 models for predicting general binding sites and active sites of proteins based on sequence alone.
-
AmelieSchreiber/esm2_t12_35M_lora_binding_sites_770K_v1
Updated • 6 -
AmelieSchreiber/esm2_t12_35M_lora_binding_sites_v2_cp3
Token Classification • Updated • 17 • 6 -
AmelieSchreiber/esm2_t12_35M_lora_binding_sites_v2_cp2
Token Classification • Updated • 10 -
AmelieSchreiber/esm2_t12_35M_lora_binding_sites_v2_cp1
Token Classification • Updated • 8
QLoRAs for various ESM-2 models for predicting binding sites of protein sequences.
This is a collection of LoRA and QLoRA finetuned ESM-2 models and datasets for predicting post translational modification sites on proteins.
ESM-2 models finetuned for generating peptide binders and predicting protein-protein interactions using MLM capabilities. None of the models work well