Upload Facade of Jade Qwen3-4B LoRA adapter

README.md

@@ -0,0 +1,43 @@
+---
+license: apache-2.0
+base_model: Qwen/Qwen3-4B-Instruct-2507
+tags:
+- qwen3
+- lora
+- peft
+- roleplay
+- build-small-hackathon
+- facade-of-jade
+library_name: peft
+---
+
+# Facade of Jade Qwen3-4B LoRA
+
+LoRA adapter for **Facade of Jade**, a small interactive wuxia dialogue game built for the Build Small Hackathon.
+
+- Base model: `Qwen/Qwen3-4B-Instruct-2507`
+- Training data: 50 curated chat examples from `loRA_training_data_v1_gpt5.jsonl`
+- Training host: Modal A100-80GB
+- Output path: `/outputs/facade-of-jade-qwen3-4b-lora`
+- Modal run: https://modal.com/apps/t-abdullah-rashid/main/ap-W54lCMfJu4eu3UCVQvVpQK
+- Final training loss: `2.969015`
+- Adapter size: `483.63 MB`
+
+## Usage
+
+```python
+from peft import PeftModel
+from transformers import AutoModelForCausalLM, AutoTokenizer
+
+base = "Qwen/Qwen3-4B-Instruct-2507"
+adapter = "build-small-hackathon/facade-of-jade-qwen3-4b-lora"
+
+tokenizer = AutoTokenizer.from_pretrained(base)
+model = AutoModelForCausalLM.from_pretrained(base, device_map="auto")
+model = PeftModel.from_pretrained(model, adapter)
+```
+
+## Project links
+
+- Live Space: https://build-small-hackathon-facade-of-jade.hf.space
+- GitHub: https://github.com/tuancookiez-hub/facade-of-jade