Fix: add useradd -m -u 1000 user + USER user (HF Spaces requirement)

Fix: use merged_model_state.pt (LoRA baked into base weights, fixes missing FFN weights)

Fix config: n_layers=8, n_monarch_heads=1, use progressive model

Upload server.py with huggingface_hub

Upload tokenizer.py with huggingface_hub

Upload requirements.txt with huggingface_hub

Upload Dockerfile with huggingface_hub

Upload server.py with huggingface_hub

Upload README.md with huggingface_hub

Upload symbio_model.py with huggingface_hub

initial commit