Update README.md

README.md

@@ -40,15 +40,20 @@ Unlike standard conversational LLMs, AegisNode Validate has been trained on a he
 Because this model relies on the native DeepSeek-Coder template, you must use the correct instruction formatting. If you download the `.gguf` file, use the following `Modelfile` to run it in Ollama:
 
 ```text
-FROM ./aegisnode-validate-6.7b-unsloth.Q4_K_M.gguf
+FROM ./aegisnode-validate-6.7b.Q4_K_M.gguf
 
-TEMPLATE """{{ if .System }}{{ .System }}
-{{ end }}### Instruction:
+TEMPLATE """{{ if .System }}{{ .System }}{{ end }}
+### Instruction:
 {{ .Prompt }}
 ### Response:
 """
 
-SYSTEM """You are a Principal AWS Cloud Architect. Write flawless, enterprise-grade Terraform code. Output ONLY valid HCL."""
+SYSTEM """You are a Principal AWS Cloud Architect. Write flawless, enterprise-grade Terraform code. Output ONLY valid HCL, no explanations or commentary."""
+
+PARAMETER stop "### Instruction:"
+PARAMETER stop "### Response:"
+PARAMETER temperature 0.0
+PARAMETER top_p 0.1
 ```
 
 Then create and run the model: