fix: connect frontend to real API endpoint

fix: derive model_dir from MODEL_PATH for config loading

fix: always use hardcoded filenames, derive dir from MODEL_PATH

fix: correct file-level skip check and config.json save path

fix: download config.json alongside model.pt

feat: HF Spaces Docker deployment setup

Override Vercel framework and build command for static deployment

Configure Vercel static deployment for frontend

Increase fine-tuning learning rate to 2e-4 in finetune_config.yaml

Fix tokenizer auto-detection in SFT and DPO training to support GPT-2 vocab size

Update kaggle_sft_train.py with automated base model finder and cleaned Alpaca downloader

Fix New Chat view update by clearing chatMessages.innerHTML

Fix LoRA unwrapping in SFT final save so output model can be served directly

Reduce max_seq_len to 512 in finetune_config.yaml to prevent OOM

Add scripts/kaggle_sft_train.py for SFT training on Kaggle

Update base_model_path in finetune_config.yaml

Fix API test compatibility with httpx 0.28+, fix model saving with tied weights, and update walkthrough documentation

upgrading the model

Support loading model.pt in from_pretrained

Support HuggingFace tokenizer in api_server.py

Add transfer.sh upload to Cell 2 in kaggle_train.py

Update kaggle_train.py with fastest working cells

Fix imports: add sys.path for Kaggle/Colab compatibility

Update pipeline.py

Update modeling_openmind.py

Simplify CI workflow by removing jobs and branches

Initial commit