Create template

`ollama run hf.co/cognitivecomputations/dolphin-2.9.4-llama3.1-8b-gguf`

To make importing with Ollama work properly, use the default chat template format.

According to instructions:
https://huggingface.co/docs/hub/en/ollama#custom-chat-template-and-parameters

template

@@ -0,0 +1,6 @@
+{{ if .System }}<|im_start|>system
+{{ .System }}<|im_end|>
+{{ end }}{{ if .Prompt }}<|im_start|>user
+{{ .Prompt }}<|im_end|>
+{{ end }}<|im_start|>assistant
+{{ .Response }}<|im_end|>