Upload 2 files

README.md

@@ -101,14 +101,21 @@ Modelfile shipped in `ollama/`:
 
 ```bash
 ollama pull hf.co/textilelabs/Loom-Spark
-ollama create loom-spark -f Modelfile     # from the ollama/ folder of this repo
+curl -L -O https://huggingface.co/textilelabs/Loom-Spark/resolve/main/ollama/Modelfile
+ollama create loom-spark -f Modelfile
 ollama run loom-spark "hi"
 ```
 
-That gives the offline persona: greetings, identity, honest deferrals. If a
-reply ends in a `<lookup>…</lookup>` line, that's the model saying *"I'd search
-for this"* — raw runners can't execute searches; for real internet answers use
-the harness (Option C).
+The Modelfile keeps multi-turn history in the trained format (each past turn is
+re-wrapped in `<user>`/`<loom>` markers) and stops generation cleanly. That
+gives the offline persona: greetings, identity, honest deferrals, made-up words.
+
+Two honest caveats: if a reply ends in a `<lookup>…</lookup>` line, that's the
+model saying *"I'd search for this"* — raw runners can't execute searches, so
+for real internet answers use the harness (Option C). And at temperature 0.85 a
+7M model occasionally misreads intent ("whats your name?" sometimes gets a
+philosophy answer; ask again or drop `--temperature 0.7`). Both quirks shrink
+in Loom Spark v2's curriculum.
 
 Note: without a wrapper that executes `<lookup>` calls and splices `<result>`
 blocks back in, GGUF/Ollama runners get the model's honest "I don't know, but

ollama/Modelfile

@@ -1,7 +1,8 @@
 FROM hf.co/textilelabs/Loom-Spark
 TEMPLATE """<tools:off>
-<user> {{ .Prompt }}
-<loom>"""
+{{ range .Messages }}{{ if eq .Role "user" }}<user> {{ .Content }}
+{{ else }}<loom> {{ .Content }}<|endoftext|>
+{{ end }}{{ end }}<loom>"""
 PARAMETER stop "<|endoftext|>"
 PARAMETER stop "<user>"
 PARAMETER stop "<result>"