Update README.md

README.md

@@ -9,200 +9,194 @@ tags:
 - unsloth
 ---
 
-# Model Card for Model ID
-
-<!-- Provide a quick summary of what the model is/does. -->
-
-
 
 ## Model Details
+**Developed by:** PranjalZetsu
+- **Model type:** Causal Language Model fine-tuned via RL (GRPO)
+- **Language(s):** English
+- **License:** Apache 2.0
+- **Finetuned from model:** [Qwen/Qwen2.5-7B-Instruct](https://huggingface.co/Qwen/Qwen2.5-7B-Instruct)
+
+- 
+| Property | Value |
+|---|---|
+| **Base Model** | Qwen/Qwen2.5-7B |
+| **Adapter Type** | LoRA (PEFT) |
+| **Quantization** | 4-bit (bitsandbytes, NF4) |
+| **Training Framework** | Unsloth + TRL (GRPO) |
+| **PEFT Version** | 0.19.1 |
+| **Task** | Text Generation / Agentic Reasoning |
+| **Domain** | Semiconductor Fabrication, Yield Engineering |
 
-### Model Description
-
-<!-- Provide a longer summary of what this model is. -->
-
+---
 
+## Model Description
 
-- **Developed by:** [More Information Needed]
-- **Funded by [optional]:** [More Information Needed]
-- **Shared by [optional]:** [More Information Needed]
-- **Model type:** [More Information Needed]
-- **Language(s) (NLP):** [More Information Needed]
-- **License:** [More Information Needed]
-- **Finetuned from model [optional]:** [More Information Needed]
+Fab_Yield_Agent_Qwen-q4 is a domain-specialized language model fine-tuned for semiconductor fabrication yield analysis and optimization. Built on the Qwen2.5-7B-Instruct architecture and trained using Reinforcement Learning from verifiable rewards (GRPO), this model learns to reason through complex statistical and materials science problems with structured, step-by-step thinking.
 
-### Model Sources [optional]
+### The Problem: Semiconductor Yield Loss
+Semiconductor manufacturing is one of the most complex industrial processes on Earth. A modern chip fab runs wafers through hundreds of steps, each controlled by multiple physical parameters (temperatures, pressures, gas flows). Even tiny deviations cascade into yield loss. This agent is trained to act as a process integration engineer:
+- Analyzing yield data and identifying root causes.
+- Navigating the physics of a 15-parameter process space.
+- Converging on optimal manufacturing recipes within a limited experiment budget.
 
-<!-- Provide the basic links for the model. -->
+### The Key Insight: RL Emergent Intelligence
+During Reinforcement Learning, the model was rewarded purely on the correctness of final answers—not on intermediate reasoning steps. Spontaneously, it developed:
+- **Deeper statistical reasoning:** Proactive use of Cpk, Poisson models, and control chart logic.
+- **Material science grounding:** Reasoning through etch selectivity, deposition uniformity, and diffusion profiles.
+- **Structured problem decomposition:** Breaking queries into logical sub-tasks before synthesizing conclusions.
 
-- **Repository:** [More Information Needed]
-- **Paper [optional]:** [More Information Needed]
-- **Demo [optional]:** [More Information Needed]
+---
 
 ## Uses
 
-<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
-
 ### Direct Use
-
-<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
-
-[More Information Needed]
-
-### Downstream Use [optional]
-
-<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
-
-[More Information Needed]
+- **Fab yield triage:** Rapidly analyze incoming yield data and identify likely root causes.
+- **Process window analysis:** Evaluate margin sensitivity across interconnected process steps.
+- **Statistical process control (SPC):** Interpret control charts and flag out-of-control signals.
+- **Material selection reasoning:** Assess tradeoffs between material properties and process compatibility.
 
 ### Out-of-Scope Use
+- General purpose QA or creative writing.
+- Safety-critical sign-off without human expert verification.
+- Use as a replacement for certified TCAD/SPC simulation software.
 
-<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
-
-[More Information Needed]
+---
 
 ## Bias, Risks, and Limitations
 
-<!-- This section is meant to convey both technical and sociotechnical limitations. -->
-
-[More Information Needed]
-
-### Recommendations
-
-<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
-
-Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
-
-## How to Get Started with the Model
-
-Use the code below to get started with the model.
-
-[More Information Needed]
-
-## Training Details
-
-### Training Data
-
-<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
-
-[More Information Needed]
-
-### Training Procedure
-
-<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
-
-#### Preprocessing [optional]
+- **Not a simulation substitute:** This model should not replace calibrated simulation tools like TCAD or certified SPC software.
+- **4-bit quantization:** While efficient, there may be minor accuracy tradeoffs compared to full-precision models.
+- **Domain Focus:** Performance on non-semiconductor tasks is not optimized.
 
-[More Information Needed]
-
-
-#### Training Hyperparameters
-
-- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
-
-#### Speeds, Sizes, Times [optional]
-
-<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
-
-[More Information Needed]
-
-## Evaluation
-
-<!-- This section describes the evaluation protocols and provides the results. -->
-
-### Testing Data, Factors & Metrics
-
-#### Testing Data
-
-<!-- This should link to a Dataset Card if possible. -->
-
-[More Information Needed]
-
-#### Factors
-
-<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
-
-[More Information Needed]
-
-#### Metrics
-
-<!-- These are the evaluation metrics being used, ideally with a description of why. -->
-
-[More Information Needed]
-
-### Results
-
-[More Information Needed]
-
-#### Summary
+---
 
+## How to Get Started
 
+### Quick Inference
+```python
+from transformers import AutoTokenizer, AutoModelForCausalLM
+from peft import PeftModel
 
-## Model Examination [optional]
+base_model = "unsloth/Qwen2.5-7B-Instruct-bnb-4bit"
+adapter    = "PranjalZetsu/Fab_Yield_Agent_Qwen-q4"
 
-<!-- Relevant interpretability work for the model goes here -->
+tokenizer = AutoTokenizer.from_pretrained(base_model)
+model = AutoModelForCausalLM.from_pretrained(base_model, load_in_4bit=True, device_map="auto")
+model = PeftModel.from_pretrained(model, adapter)
 
-[More Information Needed]
+prompt = """You are a semiconductor yield engineer.
+A 300mm fab sees 12% yield loss on a 7nm logic layer.
+Defect density: 0.08 defects/cm2. Critical area: 150 cm2.
+Perform a yield analysis and recommend corrective actions."""
 
-## Environmental Impact
+inputs = tokenizer(prompt, return_tensors="pt").to("cuda")
+outputs = model.generate(**inputs, max_new_tokens=1024, temperature=0.6, do_sample=True)
+print(tokenizer.decode(outputs[0], skip_special_tokens=True))
+```
 
-<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+### Local Setup (Environment & API)
+```bash
+# Clone the repository
+git clone https://github.com/pranjalyt/fab-yield-agent.git
+cd fab-yield-agent
 
-Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+# Install dependencies
+pip install -r requirements.txt
 
-- **Hardware Type:** [More Information Needed]
-- **Hours used:** [More Information Needed]
-- **Cloud Provider:** [More Information Needed]
-- **Compute Region:** [More Information Needed]
-- **Carbon Emitted:** [More Information Needed]
+# Run the API server
+uvicorn server:app --host 0.0.0.0 --port 7860
+```
 
-## Technical Specifications [optional]
+---
 
-### Model Architecture and Objective
+## System Architecture
 
-[More Information Needed]
+The project includes a production-grade RL environment (FabYieldEnv) and a Response Surface Model (RSM) simulator.
 
-### Compute Infrastructure
+### RSMSimulator
+A hidden ground truth engine that implements a second-order Response Surface Model. It captures non-linear interactions between 15 process parameters (Temperature, Etch Time, Pressure, etc.).
+- **Normalization:** All parameters are mapped to [-1, 1] for scale-invariant modeling.
+- **Interaction Terms:** Simulates how variables like Pressure and Gas Flow interact to control plasma density.
+- **Dynamic Physics:** Every episode generates a fresh set of coefficients, forcing the agent to generalize its optimization strategy.
 
-[More Information Needed]
+### Defect Classification
+The simulator produces physically motivated defect signatures:
+- **Edge Ring:** Non-uniform plasma at wafer edges (linked to Pressure/Gas flow).
+- **Center Spot:** Thermal hotspot at wafer center (linked to Temp/RF power).
+- **Random Scatter:** Chemical contamination (linked to Dopant levels).
 
-#### Hardware
+### Senior Engineer Reviewer
+A multi-agent layer that simulates the human approval gate. It enforces episode-varying qualification constraints (min yield, max variance, forbidden ranges).
 
-[More Information Needed]
+---
 
-#### Software
+## Training Details
 
-[More Information Needed]
+### Training Pipeline
+1. **Supervised Fine-Tuning (SFT):** Initial training on semiconductor reports, SPC problem sets, and material science Q&A.
+2. **Reinforcement Learning (GRPO):** Training via Group Relative Policy Optimization, rewarding answer correctness, structured thinking traces, and numerical precision.
 
-## Citation [optional]
+### Reward System (Four-Component Design)
+- **Yield Reward (50%):** Continuous signal based on yield improvement.
+- **Efficiency Reward (20%):** Sparse reward for hitting targets within the 12-experiment budget.
+- **Causal Attribution (15%):** Reward for correctly identifying the primary bottleneck parameter in natural language.
+- **Stability Reward (15%):** Reward for submitted recipes that show low lot-to-lot variance.
 
-<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+---
 
-**BibTeX:**
+## Evaluation: Emergent Capabilities
 
-[More Information Needed]
+### Statistical Reasoning
+| Capability | Before RL (SFT only) | After RL (GRPO) |
+|---|---|---|
+| Yield estimation | States a number | Derives from defect density + critical area |
+| Process capability | Rarely mentions | Calculates Cpk from spec limits + sigma |
+| Confidence intervals | Absent | Appears spontaneously in reasoning traces |
 
-**APA:**
+### Materials Science Reasoning
+| Capability | Before RL (SFT only) | After RL (GRPO) |
+|---|---|---|
+| Film uniformity | Generic description | Linked to deposition mechanism physically |
+| Etch selectivity | Surface-level | Reasoned from underlying chemistry |
+| Defect root cause | Names defect types | Traces cause to process physics |
 
-[More Information Needed]
+---
 
-## Glossary [optional]
+## Research Theme Alignment
 
-<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+This work addresses several frontier AI research themes for OpenEnv India 2026:
+- **World Modeling (Theme 3.1):** Modeling the partially observable professional world of a fab with physical and statistical constraints.
+- **Long-Horizon Planning (Theme 2):** Decomposing multi-step yield investigations with sparse, outcome-based rewards across a 12-step budget.
+- **Self-Improvement (Theme 4):** Emergent capability growth where the model discovers complex analytical tools to maximize its reward signal.
 
-[More Information Needed]
+---
 
-## More Information [optional]
+## Glossary
 
-[More Information Needed]
+| Term | Definition |
+|------|-----------|
+| **Wafer** | A thin silicon disc on which chips are fabricated simultaneously |
+| **Yield** | Percentage of working chips per wafer |
+| **RSM** | Response Surface Methodology - statistical technique mapping inputs to outputs |
+| **DoE** | Design of Experiments - systematic approach to planning experiments |
+| **Lot** | A batch of typically 25 wafers processed together |
+| **CMP** | Chemical-Mechanical Planarization - polishing process to flatten surfaces |
 
-## Model Card Authors [optional]
+---
 
-[More Information Needed]
+## Citation
 
-## Model Card Contact
+```bibtex
+@misc{fab_yield_agent_qwen_q4,
+  author    = {PranjalZetsu},
+  title     = {Fab_Yield_Agent_Qwen-q4: RL-Trained Semiconductor Yield Reasoning Agent},
+  year      = {2025},
+  publisher = {HuggingFace},
+  url       = {https://huggingface.co/PranjalZetsu/Fab_Yield_Agent_Qwen-q4}
+}
+```
 
-[More Information Needed]
-### Framework versions
+## Contact
 
-- PEFT 0.19.1
+For questions or feedback, open a discussion on the [Hugging Face Community tab](https://huggingface.co/PranjalZetsu/Fab_Yield_Agent_Qwen-q4/discussions).