GLM-5.3 is live. Run Z.ai's latest model on Friendli Model APIs. Try it today ➜
devanshty
Available on FriendliAI
Run this model inference on single tenant GPU with unmatched speed and reliability at scale.
Run this model inference with full control and performance in your environment.
Model Details
Model Provider
devanshty
Model Tree
Qwen/Qwen2.5-Coder-7B-Instruct
Input Modalities
Output Modalities
Supported Functionality
Code-Autopsy is a specialized code intelligence model fine-tuned on top of Qwen2.5-Coder-7B-Instruct using 4-bit QLoRA. It operates like an autonomous forensic compiler: given buggy, defective, or vulnerable code snippets across Python, JavaScript, and other languages, it outputs a clean, structured diagnostic report:
The model was trained for 3 full epochs (246 steps) on a curated dataset of code bugs and algorithmic repairs.
| Metric | Initial (Epoch 0.06) | Final (Epoch 3.0) | Delta |
|---|---|---|---|
| Training Loss | 2.162 | 0.255 | -88.2% 📉 |
Validation Loss (eval_loss) | 1.397 | 0.2442 | -82.5% 📉 |
| Token Accuracy | 60.29% | 93.20% | +32.91% 📈 |
| Gradient Norm | 0.27 | 0.39 | Stable |
🌐 Interactive Training Logs & Loss Curves:
View the live dashboard, loss charts, and hardware telemetry on Weights & Biases.
Qwen/Qwen2.5-Coder-7B-Instructbitsandbytes double quant)bfloat161632q_proj, v_projadamw_8bitpython
import torchfrom transformers import AutoModelForCausalLM, AutoTokenizer, BitsAndBytesConfigfrom peft import PeftModel BASE_MODEL = "Qwen/Qwen2.5-Coder-7B-Instruct"ADAPTER_REPO = "devanshty/Code-Autopsy" # 1. Load Tokenizer & 4-bit Base Modeltokenizer = AutoTokenizer.from_pretrained(BASE_MODEL, trust_remote_code=True)bnb_config = BitsAndBytesConfig( load_in_4bit=True, bnb_4bit_quant_type="nf4", bnb_4bit_compute_dtype=torch.bfloat16, bnb_4bit_use_double_quant=True) base_model = AutoModelForCausalLM.from_pretrained( BASE_MODEL, quantization_config=bnb_config, device_map="auto", torch_dtype=torch.bfloat16, trust_remote_code=True) # 2. Load Fine-Tuned Code-Autopsy Adaptermodel = PeftModel.from_pretrained(base_model, ADAPTER_REPO)model.eval() # 3. Format Diagnostic Promptcode_snippet = '''def append_item(val, lst=[]): lst.append(val) return lst''' prompt = f"""<|im_start|>systemYou are a code review expert. Analyze the provided code, identify any bugs or issues, explain the root cause, and provide a corrected version.<|im_end|><|im_start|>userLanguage: python ```python{code_snippet}```<|im_end|><|im_start|>assistant""" inputs = tokenizer(prompt, return_tensors="pt").to(model.device)with torch.no_grad(): outputs = model.generate( **inputs, max_new_tokens=512, do_sample=False, pad_token_id=tokenizer.eos_token_id ) print(tokenizer.decode(outputs[0][inputs.input_ids.shape[1]:], skip_special_tokens=True))The model generates responses structured in Markdown:
markdown
## Bug IdentifiedMutable default argument `lst=[]` used in function definition. ## Root CauseIn Python, default arguments are evaluated once when the function is defined, not each time it is called. Modifying `lst` mutates the single shared list object across subsequent calls. ## Fixed Code```pythondef append_item(val, lst=None): if lst is None: lst = [] lst.append(val) return lsttext
---## 📜 Citation & Credits* **Author:** Devansh Tyagi ([devanshty](https://huggingface.co/devanshty))* **Base Architecture:** Alibaba Cloud Qwen Team (`Qwen2.5-Coder-7B-Instruct`)* **Frameworks:** 🤗 Hugging Face `transformers`, `peft`, `trl`, and Weights & Biases `wandb`.## Bug IdentifiedMutable default argument `lst=[]` used in function definition. ## Root CauseIn Python, default arguments are evaluated once when the function is defined, not each time it is called. Modifying `lst` mutates the single shared list object across subsequent calls. ## Fixed Code```pythondef append_item(val, lst=None): if lst is None: lst = [] lst.append(val) return lst2e-4 (with Cosine Decay and 5% Warmup)8 (Per-device 1, Gradient Accumulation 8)