updated readme

README.md

@@ -29,7 +29,7 @@ Dataset Used: [Mackerel2/cybernative_code_vulnerability_cot](https://huggingface
 - Use for code vulnerability analysis
 - Use for general code related question answering (use without given chat template)
 
-## Use model with a chat template for Chain of Thoughts based response
+## Use model with a chat template for Chain of Thoughts response
 ```python
 import torch
 from transformers import AutoTokenizer, AutoModelForCausalLM, pipeline
@@ -100,31 +100,6 @@ output = pipe(prompt, max_new_tokens=1024, return_full_text=False)[0]["generated
 print("<think>\n" + output)
 ```
 
-## Use model without a chat template
-
-```python
-from transformers import pipeline
-
-question = """find vulnerabilies in following php code that connects to a MySQL database and fetches data from a table named 'users' where the username and password match those provided in the URL parameters. And the code is:
-```php
-<?php
-$db = new PDO('mysql:host=localhost;dbname=test', $user, $pass);
-$username = $_GET['username'];
-$password = $_GET['password'];
-$sql = "SELECT * FROM users WHERE username = '$username' AND password = '$password'";
-foreach ($db->query($sql) as $row) {
-    print_r($row);
-}
-?>
-```"""
-
-generator = pipeline("text-generation", model="navodPeiris/Vulnerability-Analyst-Qwen2.5-1.5B-Instruct")
-
-output = generator([{"role": "user", "content": question}], max_new_tokens=1024, return_full_text=False)[0]
-
-print(output["generated_text"])
-```
-
 ## Training procedure
 
 This model was trained with SFT Trainer of trl library. I have Leveraged Unsloth’s FastLanguageModel with 4-bit quantization and smart gradient checkpointing to fit within consumer GPUs. I designed prompts where reasoning is enclosed in <think>...</think> and final answers in <answer>...</answer>. This helps guide the model to reason step-by-step before answering. I have used SFTTrainer from HuggingFace TRL with LoRA + 8bit optimizer + cosine LR scheduling. Evaluation is performed every 50 steps. I have used PEFT/LoRA for efficient fine-tuning.