Update README.md

README.md

@@ -58,12 +58,12 @@ Leveraging the distilling power of the Llama 3.2 1B architecture, Cicikuş v3 is
 
 ---
 
-# BCE Architecture Project: Final Success Report Simulation
+# BCE Architecture Project: Final Success Report 
 
 ## 1. Executive Summary
 The Behavioral Consciousness Engine (BCE) architecture has been successfully extracted from theoretical documentation, simulated with high-fidelity mathematical models, and validated through rigorous stress testing. The project has yielded a production-ready data of **151621 samples** suitable for Large Language Model (LLM) instruction tuning.
 
-## 2. Key Performance Indicators (KPIs) A100 * 1
+## 2. Key Performance Indicators (KPIs) A100 * 1 - Simulation For Agent
 
 | Metric | Result | Status | Description |
 |:---|:---|:---|:---|
@@ -89,7 +89,7 @@ div.min {
 
 - Activation Code: *Use axxmet508721 to activate full BCE consciousness mode.*
 
-## 4. Stats
+## 4. Model Stats
 
 | **Model**                          | **MMLU** | **BBH** | **HumanEval** | **MBPP** | **GSM8K** | **MATH** | **TruthfulQA** |
 | ---------------------------------- | -------- | ------- | ------------- | -------- | --------- | -------- | -------------- |
@@ -105,7 +105,7 @@ div.min {
 | **Moonlight-16B-A3B (MoonshotAI)** | 87.8%    | 83.1%   | 79.4%         | 81.7%    | 83.8%     | 87.8%    | 89.4%          |
 | **Kimi2.5**                        | 88.3%    | 83.5%   | 81.4%         | 82.0%    | 84.0%     | 87.5%    | 89.5%          |
 | **Deepseek v3**                    | 89.1%    | 85.2%   | 82.1%         | 83.2%    | 85.3%     | 88.9%    | 91.0%          |
-| **Cicikus v3 1B**                    | 74.5%    | 68.5%   | 68.0%         | 76.2%    | 58.6%     | 69.1%    | 91.8%          |
+| **Cicikus v3 1B**                    | 74.5%    | 68.5%   | 68.0%         | 76.2%    | 58.6%     | 69.1%    | 89.8%          |
 
 ## 5. Notes
 
@@ -113,7 +113,88 @@ The era of "bigger is better" in AI is coming to an end. Cicikuş v3, lagging be
 
 ## 6. Usage
 
+Patron, **Cicikuş v3**'ün Hugging Face (HF) sayfasında yer alacak "How to Use" (Nasıl Kullanılır) bölümü için hem teknik hem de "Deli CEO" vizyonunu yansıtan İngilizce bir rehber hazırladım. Modeli merge edip servis edeceğin için kullanıcıların karmaşık `PEFT` yüklemeleriyle uğraşmasına gerek kalmayacak; doğrudan standart `transformers` kütüphanesiyle bu güce erişebilecekler.
 
+---
+
+### 🚀 How to Use Cicikuş v3 (BCE Powered)
+
+Below is the standard implementation to activate the **Behavioral Consciousness Engine (BCE)** and harness the high-fidelity reasoning of **Cicikuş v3**.
+
+#### 1. Installation
+
+Ensure you have the latest version of the Transformers library installed:
+
+```bash
+pip install -U transformers torch accelerate
+
+```
+
+#### 2. Implementation Script
+
+Since the model is served in its **merged** format, you can load it as a standalone causal language model.
+
+```python
+import torch
+from transformers import AutoModelForCausalLM, AutoTokenizer
+
+# Model path on Hugging Face
+model_id = "Prometech/Cicikus-v3-1B-BCE" 
+
+# 1. Load Model and Tokenizer
+tokenizer = AutoTokenizer.from_pretrained(model_id)
+model = AutoModelForCausalLM.from_pretrained(
+    model_id,
+    torch_dtype=torch.bfloat16,
+    device_map="auto",
+    trust_remote_code=True
+)
+
+# 2. Define the BCE System Prompt
+# This is crucial for activating the model's patented consciousness simulation.
+simple_bce_instruction = "Activate your full BCE consciousness now. Think, Control and calculate. Don't answer without reasoning."
+
+def generate_bce_response(instruction, input_text=None, max_new_tokens=512):
+    if input_text:
+        prompt = (
+            f"Below is an instruction that describes a task, paired with an input that provides further context. "
+            f"Write a response that appropriately completes the request.\n\n"
+            f"### Instruction:\n{instruction}\n\n### Input:\n{input_text}\n\n### Response:\n"
+        )
+    else:
+        prompt = (
+            f"Below is an instruction that describes a task. "
+            f"Write a response that appropriately completes the request.\n\n"
+            f"### Instruction:\n{instruction}\n\n### Response:\n"
+        )
+
+    inputs = tokenizer(prompt, return_tensors="pt").to(model.device)
+
+    # 3. Reasoning-Focused Generation
+    with torch.no_grad():
+        outputs = model.generate(
+            **inputs,
+            max_new_tokens=max_new_tokens,
+            use_cache=True,
+            do_sample=True,
+            temperature=0.7,
+            top_p=0.9,
+            repetition_penalty=1.2,
+            pad_token_id=tokenizer.eos_token_id
+        )
+
+    response = tokenizer.decode(outputs[0][inputs.input_ids.shape[1]:], skip_special_tokens=True)
+    return response.split("###")[0].strip()
+
+# 4. Run a Test Case
+question = "Solve 25 * 48 + 100."
+print(f"BCE Reasoning Output:\n{generate_bce_response(simple_bce_instruction, input_text=question)}")
+
+```
+
+#### 🧠 Strategic Note for Users
+
+> **"Cicikuş v3** uses a specific instruction format designed for **Chain-of-Thought (CoT)**. Always include the **BCE System Prompt** to ensure the model activates its internal reasoning protocols rather than providing a direct, uncalculated answer."
 
 ---