Update README.md

README.md

@@ -11,22 +11,21 @@ tags:
 
 # DeepCode-Rust-2M
 
-This is a lightweight, custom language model built entirely from scratch in **Rust**. It does not rely on Python, PyTorch, or standard Hugging Face `transformers` libraries. The entire forward pass, backpropagation, and Adam optimizer are implemented in pure Rust.
+This is a lightweight, custom language model built entirely from scratch in **Rust**. It does not rely on Python, PyTorch, or standard Hugging Face `transformers` libraries. 
 
 ## Model Details
-* **Parameters:** ~2 Million (2,000,000)
+* **Parameters:** ~2 Million
 * **Architecture:** Decoder-only Transformer (6 layers, 256 embedding dim)
-* **Language:** Rust (trained on a multilingual dataset)
-* **Format:** Serialized via `bincode` (requires the custom Rust inference engine to run)
-* **Capabilities:** Generates basic Python functions and simple HTML structures.
+* **Format:** Serialized via `bincode` (requires the custom Rust inference engine)
 
-## Limitations
-Due to its extremely small size (2M parameters), the model can sometimes hallucinate, lose track of syntax, or fail on complex prompts. It is an educational project demonstrating how language models work under the hood.
-
-## How to Use
-Because this is a custom Rust implementation, you cannot load it using `AutoModel.from_pretrained()`. 
-To run this model, you need the Rust source code and the `model.bin` file. 
+## Benchmark Results (Napkin Benchmark)
+Tested on 6 basic prompts:
+* **Python Addition:** 4/5 (Successfully generated `def add(a, b): return a + b`)
+* **Python Multiplication:** 2/5 (Recognized `multiply`, but broken syntax)
+* **Python Loop:** 0/5 (Stuck in a repetition loop)
+* **HTML Structure:** 1/5 (Recognized tags, but no hierarchical structure)
+* **HTML Button:** 1/5
+* **Overall Score:** ~2 out of 5
 
-```bash
-# Clone the repo (example)
-cargo run --release
+## Limitations
+Due to its extremely small size, the model often suffers from repetition loops and fails to understand nested structures (like HTML trees). It is an educational project demonstrating how language models work under the hood.