kuleshov-group/caduceus-ph_seqlen-1k_d_model-256_n_layer-4_lr-8e-3
187
Fix CaduceusForSequenceClassification scorer init
Ensure weights are tied for BiMamba (if applicable) when loaded from_pretrained
Enable mambav2 compat
Update modeling_caduceus.py
Update modeling_caduceus.py
Upload CaduceusForMaskedLM
Upload tokenizer
initial commit
