- Input: one or more protein, nucleotide, or mixed protein/DNA sequences, depending on the toolkit.
- Output: embeddings, token logits, sampled sequences, naturalness scores, gradients, or interaction maps.
Masked Models
Masked Models
Protein language models for embeddings, scoring, and sampling
Protein, coding-sequence, and mixed protein/DNA language models trained to fill in masked
positions using context from both directions. They produce embeddings, token logits,
sampled mutations, naturalness scores, and sequence gradients.
