Update README.md

README.md

@@ -63,17 +63,32 @@ The model was evaluated on a dataset containing **67,882 examples**. The evaluat
 - **Eval Samples per Second**: 7.099
 - **Eval Steps per Second**: 0.887
 
+Final performance was benchmarked using the [Mezura🥇](https://huggingface.co/spaces/newmindai/Mezura) framework — a standardized evaluation suite developed by NewmindAI for structured Turkish NLP tasks.
+
 ## Usage Example
 To use the model for **text generation** in Turkish, you can load it with the `transformers` library like so:
 
 ```python
-from transformers import LlamaForCausalLM, LlamaTokenizer
+from transformers import AutoTokenizer, AutoModelForCausalLM
+from peft import PeftModel
+import torch
+
+base_model_id = "meta-llama/Meta-Llama-3-70B-Instruct"
+adapter_id = "newmindai/Llama-3.3-70b-Instruct"
+
+tokenizer = AutoTokenizer.from_pretrained(base_model_id)
+
+base_model = AutoModelForCausalLM.from_pretrained(
+    base_model_id,
+    torch_dtype=torch.float16,
+    device_map="auto"
+)
 
-model = LlamaForCausalLM.from_pretrained("newmindai/Llama-3.3-70B-Instruct-Instruct-V3")
-tokenizer = LlamaTokenizer.from_pretrained("newmindai/Llama-3.3-70B-Instruct-Instruct-V3")
+model = PeftModel.from_pretrained(base_model, adapter_id)
 
-input_text = "Merhaba, nasılsınız?"
-inputs = tokenizer(input_text, return_tensors="pt")
-outputs = model.generate(inputs["input_ids"], max_length=50)
+prompt = "Tarhana en çok hangi il ile özdeşleşmiştir?"
 
-print(tokenizer.decode(outputs[0], skip_special_tokens=True))
+# Inference
+inputs = tokenizer(prompt, return_tensors="pt").to(model.device)
+outputs = model.generate(**inputs, max_new_tokens=100)
+print(tokenizer.decode(outputs[0], skip_special_tokens=True))