Upload README.md with huggingface_hub

README.md

@@ -10,14 +10,18 @@ suggested_hardware: t4-small
 
 # Scugnizz Llama-PCS interactive chat
 
-Custom ~1.7B **PCS** decoder (not a stock Llama). Loads `model_final.pt` from
+Custom ~1.7B **PCS** decoder (not a stock Llama). Loads weights from
 [`ProjectScugnizz/scugnizz-llama-pcs`](https://huggingface.co/ProjectScugnizz/scugnizz-llama-pcs)
-(default: **tool-context** checkpoint) and uses the same Llama-3 chat template as SFT.
+and uses the same Llama-3 chat template as SFT.
 
-Env overrides:
+**UI:** Checkpoint accordion lists Hub `training-runs/…` folders and `*.pt` files
+(`model_final.pt`, `model_pulse_*.pt`). **Load checkpoint** downloads/switches (~7GB first time).
 
-- `HUB_PATH` — training-runs/… folder (default tool-context)
+Env overrides (boot default only):
+
+- `HUB_PATH` — initial training-runs/… folder
 - `HUB_REPO` — weights repo
 - `SCRIPT_REPO` — trainer package (`scugnizz-llama.py`)
 
 Hardware: **t4-small** (~$0.40/h). Set sleep in Space settings when idle.
+Disk fills if you load many checkpoints — restart Space to clear `/tmp` caches.