Add interactive chat script with improved generation settings

chat.py

@@ -0,0 +1,56 @@
+import torch
+from transformers import AutoModelForCausalLM, AutoTokenizer
+
+print("Loading your fine-tuned Stack 2.9 model...")
+model = AutoModelForCausalLM.from_pretrained(
+    '/Users/walidsobhi/stack-2-9-final-model',
+    torch_dtype=torch.float16,
+    device_map='auto'
+)
+tokenizer = AutoTokenizer.from_pretrained('/Users/walidsobhi/stack-2-9-final-model')
+print("✅ Ready!\n")
+
+# Generation settings
+MAX_TOKENS = 150
+TEMPERATURE = 0.3
+TOP_P = 0.9
+REP_PENALTY = 1.2
+
+print(f"Settings: max_tokens={MAX_TOKENS}, temperature={TEMPERATURE}, top_p={TOP_P}\n")
+
+# Interactive loop
+while True:
+    try:
+        prompt = input("You: ")
+        if prompt.lower() in ['quit', 'exit', 'q']:
+            break
+        if not prompt.strip():
+            continue
+
+        inputs = tokenizer(prompt, return_tensors='pt').to(model.device)
+        outputs = model.generate(
+            **inputs,
+            max_new_tokens=MAX_TOKENS,
+            temperature=TEMPERATURE,
+            top_p=TOP_P,
+            repetition_penalty=REP_PENALTY,
+            do_sample=True,
+            pad_token_id=tokenizer.eos_token_id
+        )
+
+        # Extract only the new tokens (skip the prompt)
+        full_response = tokenizer.decode(outputs[0], skip_special_tokens=True)
+        response = full_response[len(prompt):].strip()
+
+        # Stop at common stop points
+        for stop in ['\n\n\n', 'You:', 'AI:', 'User:', 'Assistant:']:
+            if stop in response:
+                response = response.split(stop)[0].strip()
+
+        print(f"AI: {response}\n")
+
+    except KeyboardInterrupt:
+        print("\nExiting...")
+        break
+
+print("Goodbye!")