ESM-1b is trained from scratch by Masked Language Modeling objective. The data is combination of phosphosite data which are used to train isikz/esm1b_msa_mlm_pt_phosphosite and isikz/esm1b_mlm_pt_phosphosite.
The total number of data is 1055221.
Downloads last month
4
Safetensors
Model size
0.7B params
Tensor type
F32
·
Model tree for isikz/esm1b_pt_phosphosite_data_combined
