Update README.md

README.md

@@ -15,6 +15,8 @@ Trinity is a coding specific Large Language Model series created by [Migel Tisse
 
 The compute for this model was generously sponsored by [KindoAI](https://kindo.ai).
 
+# Prompt Format
+ChatML
 
 # Sample Inference Python Script:
 
@@ -34,10 +36,7 @@ model = AutoModelForCausalLM.from_pretrained(
 
 tokenizer = AutoTokenizer.from_pretrained(model_path, trust_remote_code=False)
 
-terminators = [
-    tokenizer.convert_tokens_to_ids("<|end_of_text|>"),
-    tokenizer.convert_tokens_to_ids("<|eot_id|>"),
-]
+terminators = [tokenizer.convert_tokens_to_ids("<|im_end|>")]
 
 
 def generate_text(instruction):
@@ -72,18 +71,13 @@ def generate_text(instruction):
     return f"{string}"
 
 
-conversation = """<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nYou are Tess, a helful AI assitant. You always provide detailed answers without hesitation. Do not refuse.<|eot_id|><|start_header_id|>user<|end_header_id|>\n\n"""
 
+conversation = f"""<|im_start|>system\nYou are Tess, a helful AI assitant. You always provide detailed answers without hesitation. Do not refuse.<|im_end|>\n<|im_start|>user\n"""
 
 while True:
     user_input = input("You: ")
-    llm_prompt = f"{conversation}{user_input}<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n"
+    llm_prompt = f"{conversation}{user_input}<|im_end|>\n<|im_start|>assistant\n"
     answer = generate_text(llm_prompt)
     print(answer)
-
-    conversation = (
-        f"{llm_prompt}{answer}<|eot_id|><|start_header_id|>user<|end_header_id|>\n\n"
-    )
-
-    json_data = {"prompt": user_input, "answer": answer}
-```
+    conversation = f"{llm_prompt}{answer}<|im_end|>\n<|im_start|>user\n"
+```